Plain English For Doctors and Other Medical Scientists 2017 PDF

i
Plain English for Doctors

and Other Medical Scientists
ii
iii
Plain English for Doctors

and Other Medical Scientists
O S C A R L I N A R E S , MD
DAV I D DA LY
G E RT R U D E DA LY
1
iv
1
Oxford University Press is a department of the University of Oxford. It furthers
the University’s objective of excellence in research, scholarship, and education
by publishing worldwide. Oxford is a registered trade mark of Oxford University
Press in the UK and certain other countries.
Published in the United States of America by Oxford University Press

198 Madison Avenue, New York, NY 10016, United States of America.
© Oxford University Press 2017
All rights reserved. No part of this publication may be reproduced, stored in

a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by license, or under terms agreed with the appropriate reproduction
rights organization. Inquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above.
You must not circulate this work in any other form

and you must impose this same condition on any acquirer.
Library of Congress Cataloging-in-Publication Data

Names: Linares, Oscar, 1957– author. | Daly, David, 1959– author. |
Daly, Gertrude, 1993– author.
Title: Plain English for doctors and other medical scientists / Oscar Linares,
David Daly, Gertrude Daly.
Description: New York, NY : Oxford University Press, [2017] | Includes
bibliographical references.
Identifiers: LCCN 2016041847 | ISBN 9780190654849 (pbk.)
Subjects: | MESH: Medical Writing
Classification: LCC R119 | NLM WZ 345 | DDC 808.06/661—dc23
LC record available at https://lccn.loc.gov/2016041847
This material is not intended to be, and should not be considered, a substitute for medical or
other professional advice. Treatment for the conditions described in this material is highly
dependent on the individual circumstances. And, while this material is designed to offer
accurate information with respect to the subject matter covered and to be current as of the
time it was written, research and knowledge about medical and health issues is constantly
evolving and dose schedules for medications are being revised continually, with new side effects
recognized and accounted for regularly. Readers must therefore always check the product
information and clinical procedures with the most up-to-d ate published product information
and data sheets provided by the manufacturers and the most recent codes of conduct and safety
regulation. The publisher and the authors make no representations or warranties to readers,
express or implied, as to the accuracy or completeness of this material. Without limiting
the foregoing, the publisher and the authors make no representations or warranties as to
the accuracy or efficacy of the drug dosages mentioned in the material. The authors and the
publisher do not accept, and expressly disclaim, any responsibility for any liability, loss or risk
that may be claimed or incurred as a consequence of the use and/or application of any of the
contents of this material.
9 8 7 6 5 4 3 2 1

Printed by WebCom, Inc., Canada
v
We dedicate this book to the doctors and other medical scientists who write. They
strive to make the world a better place by writing about new ways to understand,
prevent, treat and cure disease.
vi
vii
CONTENTS
Preface xi
Acknowledgments xvii
Introduction 1
A . W HY B O T HE R TO W R I T E IN PL A IN E NGL I SH? 1
B. W H AT D O W E ME A N BY PL A IN E N G L I S H ? 4
C . ME DICU S IN COMPR E HE N S IBIL I S 6
D. U SING T HE T IP S IN YOUR W R I T ING 7
E . NO T E S ON T HE EXE RC I SE S 10
C O N C E P T 1 ■ TAKE CHARGE OF YOUR READING EASE SCORE

A . FL E S C H R E A DING E A SE A ND FL E S C H-K INC A ID G R A DE L EV E L T E S T S 13
B. WSE G S C OR E S—H O W W E T R ACK K EY R E A DING E A SE DATA 15
CHAPTER 1. Use normal sentence length 19

A . K E EP SE N T E NC E L E NGT H 15 WOR DS AV E R AG E , 25 WOR DS M AXIMUM 19
B. K E EP T HE S UBJE CT A ND V E R B CL O SE TO G E T HE R IN T HE FIR S T SEV E N

OR E IGHT WOR DS 22
C . PU T T HE M A IN POIN T FIR S T A ND T HE N GI V E C OMME N TA RY,

DE TA IL OR S UPPORT 23
CHAPTER 2. Prefer the short word 27

A . K E EP E S SE N T I A L S C IE N T IFIC T E R MS; MINIMIZE O T HE R L ONG WOR DS 27
B. C I T E A C OMMON ME DIC A L T E R M ONC E TO AVOID C ONFU SION 29

C . W R I T E A C OMPOUND WOR D TO PROMO T E R E A DING E A SE A ND SHO W HO W
YOU PRONOUNC E I T 29
D. OMI T A N Y UNNE C E S S A RY WOR D E NDING 33

viii
viii Contents
E . AVOID T HE NOUN S T R ING 34
F. D ON’ T BE A F R A ID TO S TA RT A SE N T E NC E W I T H A ND OR B U T 36
G . AVOID U SING A HIGH PE RC E N TAG E OF L ONG WOR DS 37
CHAPTER 3. Omit any needless word 41

A . SPO T A ND OMI T NE E DL E S S WOR DS 41
B. OMI T T HE NE E DL E S S OF 42
C . OMI T T HE NE E DL E S S T H AT 43
C O N C E P T 2 ■ USE VIVID LANGUAGE
CHAPTER 4. Prefer active voice 49

A . IDE N T IFY ACT I V E A ND PA S SI V E VOIC E 50
B. R EVI SE PA S SI V E IN TO ACT I V E VOIC E 51
C . W HE N SHOULD YOU U SE PA S SI V E VOIC E? 52
D. MINIMIZE F OR MS OF T O B E A ND T O HAV E 53
E . IDE N T IFY NOMINA L IZAT ION 55
F. C ONV E RT NOMINA L IZAT ION IN TO A V E R B IN ACT I V E VOIC E 58
CHAPTER 5. Prefer concrete language 61

A . IDE N T IFY ABS T R ACT A ND C ONCR E T E S UBJE CT S 61
B. R EVI SE ABS T R ACT IN TO C ONCR E T E 64
C . U SE NOUN S A ND V E R BS TO C A R RY T HE W E IGHT OF ME A NING 65
D. W R I T E IN T HE SINGUL A R 67
E . TA LK IN T E R MS OF ONE D O CTOR T R E AT ING ONE PAT IE N T 69
CHAPTER 6. Observe the 1066 principle 73

A . PR EF E R T HE SHORT WOR D TO DE S CR IBE T HE R E A L WOR LD 75
B. PR EF E R ’S TO SHO W R E A L-W OR LD PO S SE S SION OR C ONNE CT ION 76
C . U SE T E R MS C ON SI S T E N TLY; AVOID E L E G A N T VA R I AT ION 78
D. AVOID U SING A L ONG , L AT IN WOR D TO DE S CR IBE T HE R E A L WOR LD 79
CHAPTER 7. Statistical analysis of wseg scores 85

ix
Contents ix
CONCEPT 3 ■ PRESENT LOGICAL REASONING CLEARLY
CHAPTER 8. Organize your narrative in a way that’s helpful for

your reader 93
A . IN T RODUC E A ND DEV E L OP ONE IDE A IN E AC H PA R AG R A PH 93
B. PR E SE N T T WO-D IME N SIONA L DATA IN A TABL E , C H A RT OR G R A PH 96
CHAPTER 9. Choose a clear narrative pathway 101

A . S TA RT W I T H T HINGS K NO W N 101
B. S TA RT BY A NC HOR ING YOUR DI S CU S SION IN T HE R E A L WOR LD 102
C . C HO O SE A GO OD NA R R AT I V E PAT HWAY 103
D. M AK E A SMO O T H T R A N SI T ION BE T W E E N C ONCR E T E A ND ABS T R ACT 104
CHAPTER 10. Forge a strong chain of logical reasoning 107

A . EXPL A IN E AC H S T EP OF R E A S ONING 107
B. S TAT E T HE PROBL E M BEF OR E YOU S OLV E I T 108
C . S AY I T IN WOR DS BEF OR E YOU S AY I T IN SYMB OL S 110
Afterword—Can things ever change? 115
Appendix 1. English speakers around the world 117

Appendix 2. Selected excerpts from medical science articles 125
Appendix 3. Exercise k e y 1 2 9
Glossary 201
Resources 205
Index 207
x
xi
PREFACE
Medicus incomprehensibilis is a condition that affects doctors and other medical

scientists, causing them to write dull, lifeless prose that is hard to understand. In
this book, we identify two causes: necessary scientific complexity and needless
grammatical complexity. We show how to diagnose and treat medicus incompre-
hensibilis. This involves cutting needless grammatical complexity, using proven
plain-English writing tips, while keeping essential scientific content.
Medicus incomprehensibilis wastes time for every reader, reduces understanding,
and slows the spread of new medical knowledge. This prevents medical discover-
ies from reaching the widest reasonable audience. The widest reasonable audi-
ence includes anybody with an interest in the science, including people from other
countries, and people in other specialties, sub-specialties, disciplines, or levels
of training. Treating medicus incomprehensibilis saves time for every reader and
improves understanding, thus speeding the spread of new medical knowledge.
This matters in the 21st century, when English is the global language of sci-
ence. English-language medical journals contain discoveries from doctors and
other medical scientists from around the world. Yet three out of four English
users are not native speakers. Thus, it makes sense to write about complex science
without using complex grammar.
This book is intended for doctors and other medical scientists who write for
medical science journals, and anyone who aspires to do so. It is designed for class-
room use, self-study and reference. While we focus on writing for medical science
journals, most tips also apply to other medical or science writing situations.
Who are we and how did we come to write this book?

I, Oscar, am a doctor. My parents fled Cuba when I was three years old. I grew up
in Miami speaking Spanish at home and English in school. I studied medicine in
Spanish in the Dominican Republic and earned my medical degree there.
xi
xii
xii Preface
After finishing medical school, I came back to the United States for residency.
Since I’d studied abroad, I had to take a test to qualify to work as a medical resi-
dent. Even though I knew the concepts, learning medical English was a big chal-
lenge. Luckily, both Spanish and English tend to use the same Latin and Greek
medical terms, which made things easier.
I did my fellowship in geriatric medicine at the University of Michigan, and
continued there for another 15 years as a visiting scholar. It was there I learned
medical writing through years of research and working with other doctors and
medical scientists.
Nobody ever actually taught me how to write about medicine. Instead, my
advisors told me to read articles in the journal you want to write for, study their
style, and imitate it. It was understood: You’re joining an exclusive club. You’ve
got to learn to write like members of the club.
This was hard advice to follow. Even after three years of training in internal
medicine, I had a tough time reading articles in geriatric neuroscience. They used
lots of long, formal sentences and impressive-sounding words. Many sentences
were in passive voice. Before I could understand an article, I had to study it care-
fully with Stedman’s Medical Dictionary at my side.
Whenever we wrote a scientific article, the writing process worked like this.
The lead author does the first draft, distributes it to co-authors for review, and
makes revisions. The senior author approves the final draft.
I always felt medical writing wasn’t as clear as it could be. It certainly wasn’t
reaching a wide audience. Only about 100 people in the world could understand
the articles we wrote. In fact, most of the best minds in medicine couldn’t under-
stand. A friend suggested I read Strunk and White’s book, The Elements of Style.
But, in those days, nobody in medicine seemed to know or care about writing in
plain English.
It was only when I started working with David a few years ago that I started to
see a better way to write. Together, we began to define medicus incomprehensibilis,
describe its symptoms, and explain how to treat it.
I, David, am a lawyer. I learned about plain English over the years from books,
people and lots of practice. I read Strunk and White’s The Elements of Style in
college. But nobody ever told me the ideas in that book might apply to math or
music, my two college majors.
While studying in Germany during my junior year, I learned what it was like
to be a “foreigner.” While I and the other foreign students were among the best
and brightest in every class, we always sat in the front row, straining to hear the
professor clearly. We often had minor troubles understanding ideas written in a
polite or formal German that differed from the German we spoke in everyday life.
In law school at the University of Michigan, I took the standard course on legal
research and writing, where we read Wydick’s book, Plain English for Lawyers.
I organized weekly German and Spanish conversation tables at the dining hall.
During my senior year, I taught the Introduction to American Law course for foreign
xiii
Preface xiii
legal scholars, which involved research and writing. These experiences helped me
see some of the difficulties non-natives face in reading and writing English.
Years later, I found myself at Chrysler working on international business deals.
There, I encountered many smart people, Americans and foreigners, who found
traditional contract language hard to understand.
Like most lawyers, I had learned traditional legal writing by imitating other
lawyers. When writing expert Bryan Garner came to Chrysler to put on semi-
nars on plain English for the legal staff, I was skeptical. But I tried his ideas and
found they worked. When he came back the next year, he pointed out my writing
sample as one written in almost perfect plain English. Somewhere along the way,
I also learned about Flesch Reading Ease scores. I found plain English was ideal for
international automotive deals, where people in different disciplines from all over
the world work together to build and sell cars.
After I had experience writing contracts in plain English, I wrote a few articles
for the Michigan Bar Journal. One called, “Why Bother to Write Contracts in Plain
English?”i has been cited on legal writing websites around the world. Another,
“Taming the Contract Clause from Hell: A Case Study,”ii got criticized by a reader
who thought my translation wasn’t quite right. In my response article, “The
Return of the Contract Clause from Hell,”iii I said he was right to call it a “trans-
lation.” The original was so convoluted, it might as well have been written in a
foreign language. Anything that bad was bound to be misunderstood. This whole
exchange became the subject of Prof. Joseph Kimble’s article, “The Great Myth
that Plain Language is not Precise,”iv and a chapter in his book, Lifting the Fog of
Legalese: Essays on Plain Language.v
We, Oscar and David, got the idea for this book a few years ago, when we
started working together to revise a paper on some of Oscar’s medical research.
At first, we had a clash in writing styles. But after dozens of shouting matches, we
worked everything out and the article got published. (One reviewer tried to turn
our plain-English into traditional medical writing.)
The final draft was far better than the original, and we traced the improvement
to a dozen or so principles. We thought other medical authors might benefit if we
could explain them. We started by writing down a list of rules for plain English
medical writing. One rule was, Decide what you need to say, then say it clearly and
concisely.vi This general rule applies all the time. Other rules were specific (e.g., talk
in terms of one doctor treating one patient), but only apply usually, often, or some-
times. Therefore, we thought it better to call these ideas, tips.
As we prepared to write, we checked other books on writing, found other
tips, and added them to our list. A few books we particularly liked were Anne
E. Greene’s Writing Science in Plain Englishvii and Joseph Williams’ Style: Toward
Clarity and Grace.viii As we added other people’s ideas to our own list of tips, we
noticed some ideas were more general and some, more specific. Sometimes,
different tips approached the same problem from a different angle. Some were
numerical (Keep the subject and verb together in the first seven or eight words), others
xiv
xiv Preface
grammatical or linguistic (Prefer the short word), some conceptual (Put the main
point first). We decided to include some of these different approaches, since some
readers might respond better to one approach or another.
I, Gertrude, graduated from the University of Michigan with a degree in
English. I am now a writer and editor. I run a blog about royal families with about
75% of my followers outside the USA.
Like Oscar and David, I’ve also had experience studying in another country.
During high school, I was an exchange student in Latvia. Some of my Latvian
classmates spoke English better than others, but it always came out as a Latvian
kind of English. They were obviously thinking in Latvian and translating their
thoughts. They tended to use shorter sentences, simpler grammar, and particular
words and phrases that translated well from Latvian into English. As they did
this, I found myself talking more like them so we could communicate better.
Later on, when I studied medieval, renaissance and early-modern literature at
the University of Michigan, I learned how English has changed over time. Jane
Austin, Shakespeare, Chaucer and Beowulf: the further back you go, the harder it
is for a modern audience to understand. I also learned how English has changed as
it has spread throughout the world. In this book, we urge authors to write modern
English, keeping essential scientific terms, which help explain the science, but
without archaic, exotic or obscure language that might keep some readers from
understanding.
How we structured the book

We grouped our plain-English writing tips into nine chapters. Each chapter covers
a handful of related tips. We grouped these chapters into three concepts: reading
ease, vivid language, and flow of logic.
The first concept, Take charge of your reading ease score, deals with objective mea-
sures of readability: the Flesch Reading Ease score and the Flesch-Kincaid Grade
Level. We use these scores the way a doctor uses a blood pressure cuff, to quickly
and easily get some useful information. Any good writer should know about these
scores and how to use them to help improve their writing.
The second concept, Prefer vivid language, asks the reader to ponder the differ-
ence between the real world (e.g., flesh and blood, bed pans, IV bags) and the world
of abstract ideas (e.g., diagnoses, theories, concepts). When you write, you should
never leave your reader in doubt as to which world you’re talking about.
The third concept is Present logical reasoning clearly. Logical flow is often good
for a medical journal article. But sometimes, when you fix reading ease and use
more vivid language, hidden problems with the flow of logic come to light. Once
you can see the problem clearly, it’s often easy to fix.
Another common problem is, sometimes an expert, an insider in their field,
skips steps of logical reasoning that seem “obvious” to them. Their peer reviewers,
xv
Preface xv
also experts and insiders in their field, fail to notice the missing steps. But what
is obvious to insiders may not be obvious to everybody else in the widest reason-
able audience. Often, outsiders interested in the topic could follow the train of
thought if only the expert would take care to spell out their steps of reasoning.
The tips in this book come with exercises, so you can practice what you learn.
We base most exercises on excerpts from six leading medical journals: Journal
of the American Medical Association, The Lancet, New England Journal of Medicine,
British Medical Journal, Mayo Clinic Proceedings, and American Family Physician. We
chose these journals since they represent today’s best “standard” medical writing.
Writing in plain English is a skill you must hone by practice. Our goal is to
point you in the right direction, give some general guidance and specific tips, and
then, encourage you to develop your own creativity and good judgment. For most
exercises, we conclude by asking you to make any other changes you can think of to
improve reading ease.
For the exercises in Chapters 1–6, we check reading ease scores for the original
excerpts and our revisions in the Exercise Key. In Chapter 7, we analyze these data
to show how reading ease improves.
The tips in this book are specific, easy to use, and they really work. Many are
simple (e.g., Use normal sentence length). Yet for each tip, we found examples in
journals where the authors did the opposite, and the editors and peer reviewers
let it pass.
Medical science writing is important. It is every medical writer’s duty to write
clearly and concisely. Learning to express complex ideas clearly and concisely is in
no way a remedial skill. Rather, it can only been seen as a sign of mastery.
Notes
i. Daly D, “Why Bother to Write Contracts in Plain English?” Mich Bar J, (August 1999): 850–851.
ii. Daly D, “Taming the Contract Clause from Hell: a Case Study,” Mich Bar J, (October 1999):
1155–1157.
iii. Daly D, “The Return of the Contract Clause from Hell,” Mich Bar J, (February 2000): 202–204.
iv. Kimble J, “The Great Myth That Plain Language Is Not Precise,” Scribes J Leg Writing
7 ( 1998–2000 ), 109.
v. Kimble J, Lifting the Fog of Legalese: Essays on Plain Language (Durham, NC: Carolina Academic
Press, 2006): 37–48.
vi. This rule came from one of Bryan Garner’s live seminars.
vii. Greene A, Writing Science in Plain English, (Chicago: University of Chicago Press, 2013).
viii. Williams J, Style: Lessons in Clarity and Grace, 9th ed. (New York: Pearson Longman, 2007).
xvi
xvii
ACKNOWLEDGMENTS
We thank and acknowledge several people for their contributions:

For their years of mentorship and collaboration in writing about medical science
and mathematical modeling: Raymond C. Boston, MSc, PhD, Abramson Cancer
Center of the School of Medicine, Professor Emeritus of Biostatistics, Department
of Biostatistics and Epidemiology, Perelman School of Medicine, University of
Pennsylvania; William E. Schiesser, PhD, ScD, Emeritus McCann Professor of
Chemical and Biomolecular Engineering and Professor of Mathematics at Lehigh
University; Jeffrey B. Halter MD, Professor of Internal Medicine and Director,
Geriatrics Center at the University of Michigan; and Loren A. Zech, MD, Senior
Scientist, National Institutes of Health.
Bryan A. Garner for his live seminars on plain English writing at Chrysler
Corporation and his many fine books on plain English.
Joseph Kimble, distinguished professor Emeritus, Western Michigan
University—Cooley Law School, for his guidance and encouragement as editor of
the plain English column of The Michigan Bar Journal, and as President of Clarity
International.
Anne E. Greene for her book, Writing Science in Plain English, which in many
ways served as a model for this book.
Miriam S. Daly, MD and Mary Laur for their review and comments on early
drafts.
Janice L. Bernick, PhD, FNP; John J. Bernick, MD, PhD; William E. Maxwell,
Jr., JD; and Annemarie L. Daly, MD, JD, FACP for their support and encourage-
ment during the time we wrote this book.
Andrea Knobloch, Allison Pratt, Emily Perry, Cheryl Jung, Christopher Reid,
Lani Oshima, and the other wonderful editors and staff at Oxford University
Press who helped bring this book to print.
xvii
xviii
1
Introduction
Since writing is the only means the medical profession has of universally
disseminating knowledge concerning new therapeutic concepts, medical
discoveries, or clinical experience, it is the moral obligation of every
physician who has made an original scientific observation or has for-
mulated from his own experience a new medical theory, to publish
it for the information of his colleagues, and the ultimate benefit of
mankind.—Selma DeBakey1
A. Why bother to write in plain English?

Good writing takes work. If your article is good enough to get published, why
should you make an extra effort to ensure it is clear, concise and readable? Here
are ten reasons.
1 . I T HE LP S S PR E A D N EW ME DIC A L K NO W L E DG E
When doctors share ideas about theory and practice, it helps spread medical
knowledge. New medical discoveries prevent illness, relieve suffering, find cures,
and extend life. Difficult-to-understand writing slows down this process; clear,
concise writing speeds it up.
2 . I T HE LP S T E AC H T HE PROF E S SION
The Hippocratic Oath is a tradition of the medical profession; many doctors take it
when they graduate from medical school. As part of the oath, they pledge to teach
others the profession “according to their ability and judgment.” This implies they’ll
do what they can to make medicine understandable to others.
3 . I T S HO W S R E SPE CT F OR T HE R E A DE R
Doctors are busy professionals. When you write in plain English, it shows respect.
Your reader will read faster, understand better, and remember longer. Research in
1
2
2 Plain English for Doctors and Other Medical Scientists
other fields shows 80% of readers prefer plain English.2 Even if they can under-
stand an article written in a traditional style, doctors are human too. Their brains
work the same way as everybody else’s. The same factors that influence reading
ease for everybody else also influence reading ease for doctors.
4. I T S AV E S R E A DIN G T IME
How much time does the average doctor spend reading medical journals? Research
published in the Journal of the Medical Library Association reports the average
pediatrician spends 118 hours a year.3 The Journal of General Internal Medicine
reports the average internist spends 228 hours a year.4
What is the potential savings in reading time? Since there is little research on
plain English medical writing, we can only make an educated guess based on expe-
rience in other fields. Writing expert Robert Eagleston thinks writing in plain
English may cut reading time by 30% to 50%.5 Joseph Kimble tested traditional
and plain English contracts on various groups of readers and found that plain
English cut reading time between 4.7% and 19.7% while improving comprehen-
sion.6 Considering how much doctors read and how much time plain English
saves, it seems likely that, if all medical journals were written in plain English, it
could save the average doctor a week or two per year.
Writing well may take extra work, but even if it does, it’s worth it. If your arti-
cle is published and circulated to thousands of readers, any extra time you spend
to write more clearly and concisely is small compared to the total time you save
for your readers.
5. I T HE LP S YO U R WOR K R E AC H T HE W IDE S T

R E A S ONAB L E AU DIE N C E
When you consider the widest reasonable audience, and write in a style suitable for
them, it promotes free and efficient exchange of new medical knowledge. Ideally,
a medical journal article should be accessible to anybody interested in the sub-
ject matter, whether or not they are an insider in a field. This may include a doc-
tor or scientist working in the same or a related specialty, a student, or a nurse.
It includes regular journal subscribers and those who search for articles on the
internet.
English is the global language of science, but many people who read English-
language medical journals are not native speakers. In fact, according to English-
language expert David Chrystal, non-native English speakers outnumber native
speakers 3:1.7 The widest reasonable audience includes those living or educated in an
English-speaking country or elsewhere, whether a native speaker of English or not.
In Appendix 1, we present a survey of English speakers around the world. Some
of the conclusions we reached in putting it together surprised us. For example, did
3
Introduction 3
you know there are now 23 countries where more than 10 million people speak
English? Did you know there are more people who speak English in Asia than in
North America? And did you know Germany is now the seventh largest English-
speaking country in the world with more than 50 million English speakers? Many
German doctors read English-language medical journals. They may have a first-
rate medical education but only read English at a USA high school level. Given
that English-language medical journals are read by doctors throughout the world,
it makes good sense to write complex science in plain English.
6 . I T HE LP S PL E A D F OR A C AU SE
Medical science articles advocate for people who are poor, sick or oppressed. If you
don’t make your point clearly, you’re not helping anybody. Writing in an inflated
formal style sends the message there’s no urgent problem. It’s just business as usual
for those of us who work in a hospital, university or research center. Writing in plain
English helps send the message that a problem is important and urgent.
7 . I T HE LP S H UM A N IZE YOUR W R I T IN G
Writing in plain English sounds more natural, closer to the way people speak in
everyday life. It sounds professional, with careful attention to the science, but less
formal and bureaucratic. This means your writing sounds more human and puts
less of a burden on your reader.
8 . I T S HO W S R E SPE CT F OR YOUR WOR K
Your research and ideas deserve to be presented clearly and concisely. Good writ-
ing helps build your reputation and benefits your career. More people will read
your work and want to work with you.
9 . I T HE LP S OV E R C OME E DI TOR I A L BL I NDNE S S
Writing in plain English helps you overcome editorial blindness, that feeling you
get when you work on an article so long you miss problems a reader with a fresh
eye would see. The tips in this book help you look at your writing from a fresh
perspective.
1 0 . I T S AV E S T IME A N D IMPROV E S C ON T E N T
As a medical researcher, your time is valuable. Learning to write and revise in

plain English may take some extra time and effort initially; but, once you learn
how, it saves time for everybody. So ultimately, there is no extra cost.
4
Most medical journal articles involve multiple authors and go through peer
review. This involves many people and many steps. Getting the first step right, by
making sure your draft is clear and concise, saves time and effort at each later step.
Let’s look at the process from the lead author’s point of view. You start by writ-
ing the first draft as clearly and concisely as you can. Next, your co-authors review
the draft. Since the draft is clear and concise, they focus more on content and less
on form. They read faster, understand better, and give better comments. They
become more personally invested in the work, and team morale improves.
Once you get comments, you incorporate them into a revised draft. Since the
original was clear and concise, this revision goes quickly. Since you got better
comments from your co-authors, the content gets better. Since you continue to
check reading ease as you revise, the revised draft becomes even clearer and more
concise.
Now, your co-authors review the second draft, which reflects everybody’s com-
ments. They see the improved content and this lifts team morale further. They
become more invested in the research effort. You and your co-authors continue to
review and revise as long as you need.
Once you’re satisfied with the article, you submit it to a journal. If the journal
likes it, they assign it for editorial and peer review. Just as with co-author review,
this review goes faster and generates better comments.
Writing a first draft in plain English may take longer, but, after that, each step of
review and revising goes faster and helps improve clarity, conciseness and content.
As a result, you end up with a better paper, possibly with no more time and effort.
You can’t learn to write in plain English just by reading a book. You must put
your own pen to paper.8 This is why we included exercises for each chapter.
Exercise A. Widest reasonable audience

Look at Appendix 2, Excerpt 1 from Mathematical Modeling of Kidney Transport:
Glomerular Filtration.
1. What types of people might want to read an article about modeling how a
kidney filters blood? In other words, who is the widest reasonable audience?
2. What changes would you suggest to make the article clearer for different peo-
ple in the widest reasonable audience? (E.g., a mathematician might need a
better explanation about kidney anatomy.)
Compare your answers to the Exercise Key in Appendix 3.
B. What do we mean by plain English?

When we talk about plain English, we mean writing that conveys the right con-
tent, clearly and concisely. Some people think writing in plain English involves
5
Introduction 5
dumbing down medical science, but this is not so. Rather, it involves sharpening
up the science to make it clearer and more accessible to the widest reasonable
audience.
By content, we mean essential scientific content, those important scientific
ideas an author must include in their article. What content is non-essential?
Sometimes, an author loads down a sentence with asides and parentheticals that
only loosely relate to the main idea. Only an author can judge what content is, or
isn’t, essential. An author should never sacrifice essential content to make their
article easy to read, but they might cut non-essential content.
Writing is clear when the narrative uses words and concepts familiar to the
reader. Ideally, a reader can understand and vividly imagine the article on first
reading without having to study it. The reader remembers each key idea.
Writing is concise when it demands as little of the reader’s mental energy as
possible. This usually means short while still clear. Good writing involves trade
offs. A few short words may convey the message more vividly than one long but
lifeless word. Writing concisely means cutting an unnecessary word, but cutting
too many words may make the message cryptic and harder to understand. For
example, a math equation is very concise, but a reader might understand the same
idea better if the author explained it in words first.
Figure I-1 presents a diagram that shows how we think about plain English.
Within the universe of all the possible ways you might write a medical journal
article, plain English represents the intersection of three ideas: (1) essential sci-
entific content (2) presented clearly and (3) concisely. This diagram represents
plain English—what we are trying to achieve. We devote the rest of this book to
explaining how to achieve it.
Essential
scientific
content
Wordy Cryptic
Plain
English
Dumbed
Clear Concise
down
Figure I-1 Plain English is essential scientific content, presented clearly

and concisely.
6
Exercise B. How does sentence length affect reading ease?

Appendix 2 contains excerpts from three articles from medical science journals.
Read the first paragraph of each out loud.
1. Count the number of words in each sentence and compute the average sen-
tence length. Which excerpt uses the longest sentences? The shortest?
2. Which excerpt do you find easiest to read? The hardest?
3. Judging from these excerpts, do you see any link between reading ease and
sentence length?
C. Medicus incomprehensibilis

Open any leading medical journal today and you’ll find an article that shows
symptoms of medicus incomprehensibilis, a condition that affects doctors and
other medical scientists, causing them to write dull, lifeless prose that is hard to
understand. Medicus incomprehensibilis wastes time for every reader. It conveys
the scientific message less clearly, vividly and memorably. It keeps some readers
from fully understanding. It causes others to give up, since reading becomes too
tedious. It excludes any potential reader who might understand the science, but
can’t deal with the complex writing style.
Can we explain medicus incomprehensibilis just by the inherent complexity of
medical science? No. Certainly, in order to understand a medical journal article, a
reader must know something about biology, chemistry, physics, anatomy, epide-
miology, pharmacology, mathematics and statistics. But in many cases, complex
science is a distant secondary cause of medicus incomprehensibilis. The primary
cause is a needlessly complex traditional medical writing style, which overuses
several writing habits. See Table I-1. Any one of these symptoms is not necessarily
Table I-1. Symptoms of medicus incomprehensibilis: over-used writing habits
Low reading ease Abstraction Unclear logical reasoning

• long sentence (>25 words) • abstract language • long paragraph
• run-on sentence • formality • awkwardly organized
• dependent clause • nominalization information
• parenthetical statement • plural subject • unclear paragraph topic
• long word • obscure jargon • unclear narrative pathway
• unnecessary word ending • missing step of logical
• noun string reasoning
• passive voice
7
Introduction 7
Table I-2. Exercise C. Symptoms of medicus incomprehensibilis observed
Excerpt 1 Excerpt 2 Excerpt 3

Low reading ease ✓ ✓ ✓
long sentence (>25 words)
run-on sentence
dependent clause
parenthetical statement
long word
passive voice
Abstraction ✓ ✓ ✓
abstract language
nominalization
formality
plural subjects
obscure jargon
a problem. But the more symptoms a piece of writing has, the worse the case of
medicus incomprehensibilis.
Exercise C. Recognizing symptoms of medicus incomprehensibilis

Look again at the excerpts from the three journal articles in Appendix 2. Read the
first paragraph or two of each and make a checkmark in Table I-2 to indicate each
symptom of medicus incomprehensibilis you observe.
D. Using the tips in your writing

The tips in this book can serve as a checklist to help you write and revise more effi-
ciently. For example, if you take care to follow the tip, Use normal sentence length,
you will never waste time writing and revising a 50-word sentence. If you read
the tips and do the exercises, you will learn how to write medical science in plain
English. As you apply the tips, your writing will start to improve right away. With
steady practice, it will continue to improve.
“ T I P S ” —N O T RUL E S
Everybody who writes a book about writing style gives a list of rules. In develop-
ing our list, we checked other books and articles and tried to include all the best
ideas that apply to medical writing. (See Resources.)
8
Like medicine, writing is both an art and a science. It involves many choices
about how to say things. There is only one rule for plain English medical writ-
ing: Decide what you need to say, then say it clearly and concisely. This rule applies all
the time.
We call the other suggestions in this book tips. Most are specific and easy to
apply. We expect an author to use good judgment about when a tip may prove
helpful. We offer each tip with the provisos: “when it helps,” “when it works,”
“when it makes sense,” and “if it isn’t confusing, distracting, or vulgar.” For exam-
ple, one tip is, Prefer active voice. Many medical writers use passive voice too much
and their reading ease and vividness suffer as a result. But, if you use passive voice
sparingly, it can work well.
Think of each tip like a tool in your toolbox of writing skills. Like any tool,
not every tip helps in every situation. A hammer doesn’t help much to fix a
clogged sink. If a tool doesn’t work in a particular situation, that doesn’t mean
you throw it away. Instead, you keep it in your toolbox for when it serves a use-
ful purpose.
PL A I N E NG L I S H W R I T IN G T IP S HE LP IMPROV E A L L K I NDS
OF W R I T I N G
Many medical journal articles suffer from medicus incomprehensibilis; some are
worse than others. Some are almost unreadable. Some are tedious. Some are
readable, but take study. Still others are easy to read, but wordy. In this book, we
address each of these issues. The tips we provide can help make poor writing good,
and good writing better. Don’t stop improving just because you feel your writing
is good enough.
S M A L L C H A N G E S A DD UP
Some of the writing tips we present in this book may seem trivial, but many
small changes quickly add up to a big improvement in reading ease and clarity.
Each individual tip may only improve your article’s reading ease score by a few
points, make one idea a little more vivid, or improve the flow of logic in a small
way. But by making many small changes, you may see your reading ease score
jump, a hazy idea become clear, or a weak chain of logical reasoning become
strong.
K A I Z E N (改善)
Japanese quality management principles used in industry have strongly influ-

enced our way of writing about medicine. Kaizen, the Japanese term for improve-
ment or self-changing for the best of all, refers to a philosophy and a set of practices
that focus upon improving any process. By improving processes, kaizen aims to
9
Introduction 9
eliminate waste. Kaizen is a daily process, which goes beyond improving produc-
tivity. It is also a process that, when done correctly, humanizes the workplace,
eliminates overly hard work (muri), and teaches people how to spot and elimi-
nate waste in business processes. While kaizen usually delivers small improve-
ments, the culture of continual small improvements yields large productivity
improvements.9
C H A L L E N G I NG C ON V E N T ION
Some of our tips differ from traditional notions about “proper” medical writ-
ing. We challenge any convention that serves no scientific purpose, yet makes
an article needlessly complex or abstract. If you ever feel one of our tips violates
a convention of the profession, ask yourself, Is this convention based on a scien-
tific reason, or a social reason? If the reason is scientific, by all means, follow the
convention.
But if the convention is based on a social reason, ask yourself, How impor-
tant is that social reason? Is it more important than respecting a reader’s time? Is it
more important than making an idea accessible to the widest reasonable audience?
The benefits of plain-English outweigh any superficial notion about “proper”
writing.
You might ask, Aren’t medical writing habits that contribute to medicus incompre-
hensibilis of long standing? Not so much as some people think. Many medical writ-
ers of the past wrote their scientific works clearly. For example, Edward Jenner’s
1798 paper, “An Inquiry into the Causes and Effects of the Variolae Vacciniae,
or Cow-Pox,” which first coined the term vaccination,10 was written for a wide
audience and does not overuse difficult scientific language. Watson and Crick’s
1953 article, “Molecular Structure of Nucleic Acids: A Structure for Deoxyribose
Nucleic Acid,” which first proposed the double-helix structure of DNA, is short
and readable.11
S TAG E S OF G R IEF
If you’ve spent years learning to write in the traditional medical writing

style, you may feel a deep sense of grief and loss in reading this book, since
it reveals weaknesses of the traditional style. If so, you may go through some
of the stages of grief Elisabeth Kübler-Ross described in her book, On Death
and Dying. These stages of grief are denial, anger, bargaining, depression, and
acceptance.12 We worked through some of these same feelings in planning and
writing this book.
If you never learned to write in the traditional medical style, so much the
better. We hope this book spares you any such feelings of grief and loss, but
we also urge you to be patient with those who are going through the grieving
process.
10
E. Notes on the exercises

Anybody who wants to learn to prevent, diagnose and treat medicus incomprehen-
sibilis must spend some time revising writing samples. The exercises give you a
chance to practice applying the tips to excerpts from published journal articles.
We give our own revisions in the Exercise Key in Appendix 3.
You may feel uneasy trying to rewrite somebody else’s work. Like dissecting a
cadaver, this may seem unpleasant, but it is a part of learning.
Don’t worry if your revision differs from ours. Each revision reflects our under-
standing of the excerpt taken out of context. We had two MD’s review each exer-
cise. But in some cases, we weren’t sure what the original was trying to say, even
after careful study. We don’t claim our revision is the best possible, or that it’s
better than the original.
Don’t interpret the fact we chose any excerpt as criticism. In some cases, we
just wanted a sample of medical writing for you to practice on.
Most exercises focus on one tip, but also ask you to make any other changes
you can think of to improve reading ease. This is how revising works: you set out
to fix one thing, and as you do, you see other things to fix. Our revisions reflect
applying many tips, not just the one featured in the exercise.
Somebody might say one of our revisions isn’t entirely correct, changes the
meaning or sense of the original, over-simplifies something, or loses some essen-
tial scientific content. Despite our best efforts, we know this may be so. If a
revision doesn’t involve the author, it’s hard to be sure whether it is correct or
captures the sense of the original. If we made any mistake, or if anybody inter-
prets a passage in a different way, perhaps the original wasn’t clear. In any event,
losing key content is never a problem for an author revising their own work. Our
goal is to teach you to revise your own work, not to translate other people’s work.
Conclusion
If you’ve read this far, you’ve already taken the biggest step toward writing better.
You accept that it’s possible to write about medicine in plain English and that it’s
worthwhile. Most importantly, you’re willing to try.
Notes
1. DeBakey S, “Suggestions on Preparation of Medical Papers,” JAMA 155, no. 18 (1954): 1573.
2. Kimble, Lifting the Fog of, 3–13 (see Preface, n. 5).
3. Tenopir C, et al. “Journal Reading Patterns and Preferences of Pediatricians,” J Med Libr Assoc 95,
no. 1 (2007), under “Background,” http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1773049/.
4. Saint S, et al. “Journal Reading Habits of Internists,” J Gen Intern Med 15, no. 12 (December
2000), under “Reading Habits,” http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1495716/.
11
Introduction 11
5. Collins L, “Plea for Plain English,” QNP (December 9, 1993): 33.

6. Kimble J, “Answering the Critics of Plain Language,” Scribes Journal of Legal Writing 5
(1994–1995): 70–71.
7. Crystal D, English as a Global Language, 2nd ed. (Cambridge: Cambridge University Press,
2003), 69.
8. Wydick R, Plain English for Lawyers (Durham, NC, Carolina Academic Press, 2005): 6.
9. Wikipedia, s.v. “Kaizen,” https://en.wikipedia.org/wiki/Kaizen (accessed August 9, 2015).
10. Jenner E, “An Inquiry into the Causes and Effects of the Variolae Vacciniae, or Cow-Pox,” in The
Harvard Classics, ed. Charles W. Eliot (New York: Collier) 38: 153–180.
11. Watson J, Crick F, “Molecular Structure of Nucleic Acids: A Structure for Deoxyribose Nucleic
Acid,” Nature 171, no. 4356 (April 1953): 737–738.
12. Kübler-Ross E, On Death and Dying (New York: Scribner, 2003).
12
13
concept
1
TAKE CHARGE OF YOUR
READING EASE SCORE
Everything that can be thought at all can be thought clearly. Everything
that can be said can be said clearly.—Ludwig Wittgenstein1
In the first three chapters, we discuss how you can improve your writing by
improving your reading ease scores. We address reading ease first, since it repre-
sents the “low hanging fruit” of writing in plain English.
A. Flesch Reading Ease and Flesch-Kincaid

Grade Level tests
Flesch Reading Ease is a readability test that indicates how difficult it is to read a
passage in English. The scores generally range from 0.0 to 100.0. Higher scores
indicate material that is easier to read; lower scores, more difficult. It’s possible
for a passage to have a negative score.2
The Flesch–Kincaid Grade Level, a related test, assigns a USA school grade level
to a reading passage. Both tests were developed by the United States Navy as a
way to assess how difficult a passage of text is to read. Both tests use the same core
measures, word length and sentence length, but use different weight factors. The
tests are objective, since they consider only form and not content.3,4 The results
of the two tests correlate roughly inversely. A text with a high reading ease score
should have a low grade level score. Table C1-1 shows an approximate correlation
of the reading ease and grade level scores.
A writing sample is generally considered to be in plain English if it falls in the
60–70 range for reading ease and has about an 8 or 9 grade level.5 What scores
14
Table C1-1. Flesch Reading Ease vs. Flesch-Kincaid Grade Level1
Reading ease score Grade level score

90 to 100 5th grade
80 to 90 6th grade
70 to 80 7th grade
60 to 70 8th and 9th grade
50 to 60 10th to 12th grade (high school)
30 to 50 college/university (13 to 16)
0 to 30 college/university graduate (17 and above)
1
Flesch, “How to Write in,” (see Concept 1, n. 4).
should we expect for plain-English medical writing? Because medical writing uses
Latin or Greek names and other scientific terms, we think a good reading ease
score tends to be a bit lower (about 46–70), and the grade level a bit higher (about
7–11). We show in Chapter 7 how we came up with these ranges.6
W HY WOR RY AB OU T R E A DIN G E A SE?
The Flesch Reading Ease and Flesch–Kincaid Grade Level tests give a good gen-
eral idea of how difficult it is to read a passage of text. They are also quick and
easy to calculate. Chances are your computer’s spell checker can calculate them
for you.
We use these tests the way a doctor uses a blood pressure cuff, to get some
useful information quickly at low cost. We often use them in a way that is “off
label,” to check just one sentence or paragraph rather than a whole article. But,
the shorter the block of text, the more unreliable the score may be in reflect-
ing actual reading ease. Some things are probably not so hard for a doctor to
read as the score might suggest (e.g., names of body parts, medicines, surgical
procedures).
W HE R E C A N YO U FIN D T HE SE T E S T S?
The Flesch Reading Ease and Flesch-Kincaid Grade Level tests come as a standard
feature of the MS Word spelling and grammar checker. They are also bundled with
other popular word processing programs and services. (You may need to enable
style checking on your word processing program. Check the program’s website
for user help.) Different versions may provide slightly different scores. If your
word processing program doesn’t have these tests, you can find them free online
(https://readability-score.com).
15
Concept 1: Take Charge of Your Reading Ease Score 15
SHI F T I NG T HE CUR V E F OR R E A DIN G E A S E S C OR E S
Journal articles written in a traditional medical writing style naturally have a

range of reading ease scores. The scores for any individual article depend upon
the article’s science content and the author’s writing style. We might draw a curve
to represent the range of reading ease scores for different journal articles, some-
thing like Figure C1-1. (See left side—“ Traditional.”) Here, we represent the mean
Flesch Reading Ease score for traditional medical writing as “µT.”
By taking care to improve reading ease scores, it ought to be possible to shift
the entire curve to the right. (See right side—“Plain English.”) After this shift, the
reading ease scores will still vary, depending upon each article’s science content
and the authors’ writing style. We represent the mean reading ease score for plain
English medical writing as “µPE.”
B. wseg scores—How we track key reading ease data

We find four items of data helpful to assess reading ease for a passage of text: the
number of words (w), average sentence length (s), Flesch Reading Ease score (e),
and Flesch-Kincaid Grade Level (g). We abbreviate these four items of informa-
tion as “wseg.” For example, if a paragraph has 36 words, an average sentence
length of 18.0 words, a reading ease score of 71.0, and a grade level of 7.8, we
would summarize this information as (wseg = 36/18.0/71.0/7.8). We use wseg
scores to track changes in reading ease throughout this book.
Example—Using reading ease scores to help diagnose and treat medicus

incomprehensibilis
It is possible to detect symptoms of medicus incomprehensibilis even in garden-
variety traditional medical writing. For example, here is a sentence from an
Traditional Plain English

Mean reading ease = μT Mean reading ease = μPE
Frequency
0.0 μT μPE
Flesch Reading Ease (FRE) score
Figure C1-1 Plain English can materially improve a journal’s reading ease.
16
article in the American Family Physician, a journal read by 190,000 family practice
doctors.
Referral to a clinical geneticist for assistance in the diagnosis and man-

agement of Noonan syndrome, including determining the appropri-
ateness and sequence of genetic testing, may be helpful.7 (wseg = 27/
27.0/0.0/20.7)
We understand this sentence, but it’s not as easy to read as it might be. An occa-
sional sentence like this doesn’t pose a problem for the average medical journal
reader. But when every sentence sounds like this, an article soon becomes tedious
and fails to serve the widest reasonable audience.
The wseg score gives us a few hints about this sentence. First, it is a little
long: one sentence of 27 words. The reading ease score is 0.0. The grade level is
20.7, which may not pose a problem for a doctor educated in English. But it might
for a doctor educated in another language.
What makes this sentence challenging to read—the science or the writing
style? Certainly, understanding the term, Noonan syndrome, requires some medi-
cal knowledge. But, in this case, the whole article is about Noonan syndrome; the
reader will likely learn what it involves. A reader can also tell Noonan syndrome is
a medical condition. Thus, the science is a not the cause for this sentence’s low
reading ease.
As we see it, the low reading ease springs from commonly overused writing
habits:
• Many words come between the subject (referral) and the verb (may)
• A long sentence—27 words
• A long dependent clause in the middle (including determining the appropriate-
ness and sequence of genetic counseling)
• Abstract nouns: referral, assistance, diagnosis, management, appropriateness and
sequence
• Slight redundancy—determining the sequence of genetic testing is part of
determining appropriateness
If we try to revise the example to keep the same scientific content, but avoid or
minimize these writing habits, we might end up with something like the revision
in Table C1-2.
In this revision, the number of words is the same, but since the words are
shorter, it takes up less space on the page. Since it uses two sentences, the aver-
age sentence length is one-half. The Flesch Reading Ease score jumps from 0.0
to 61.5. The grade level goes down from 20.7 to 8.0. Clearly, if every sentence
17
Concept 1: Take Charge of Your Reading Ease Score 17
Table C1-2. Revising to reduce medicus incomprehensibilis
Original Revised
Text Referral to a clinical geneticist It may help to refer a patient with
for assistance in the Noonan syndrome to a clinical
diagnosis and management geneticist. They can help diagnose
of Noonan syndrome, and manage it and help decide on
including determining the genetic tests. (wseg = 27/13.5/
appropriateness and sequence 61.5/8.0)
of genetic testing, may be
helpful.
(wseg = 27/27/0.0/20.7)
Words 27 27
Average words
27 13.5
per sentence
Reading ease
0.0 61.5
score
Grade level 20.7 8.0
in every article could be revised like this, a journal could serve a much wider
audience.
Is this the best possible revision? Maybe you can think of something better. Or
maybe you would have made some different revising decisions than we did and
achieved a different wseg score. The point is, when you focus on reading ease, you
can often make a big improvement.
Exercise D. Take charge of your reading ease score
1. Read the following excerpt out loud and underline each word with three syl-
lables or more (but skip expanded and produces, which are two-syllable words
that add common endings).
2. For each word you underlined, can you think of a shorter word or a few short
words that mean close to the same thing? Or else, do you consider it an essen-
tial scientific term, not easy to replace?
3. Revise the excerpt to break up the long sentences into shorter ones. Keep any
essential scientific term, but use shorter words where you can. Do you find
your revision easier to read than the original?
18
Given the complexity of the immune system, the development of

an individual B cell is unlikely to follow a predictable and well-
executed series of decision points whereby an antigen-reactive cell
is expanded to a clone that produces a single antibody. A more real-
istic perspective is that their development depends on a series of
error-prone, random rearrangement events and mutations whereby
specificity for the original antigen is maintained (or not) by selec-
tive pressures.8
Conclusion
The first concept is Take charge of your reading ease score. A reading ease score is
based on two factors: average sentence length and average word length. In the
next three chapters, we give tips for improving reading ease scores:
1. Use normal sentence length

2. Prefer the short word
3. Omit any needless word
Notes
1. Williams, Style: Lessons in Clarity, 3 (see Preface, n. 8).
2. For the exercises in this book, negative scores are truncated to zero. This is a standard feature
for the MS word style checker. We present an analysis of the wseg scores for excerpts used in
Chapters 1–6 exercises and our revisions in Chapter 7. For this analysis, we calculated negative
scores manually.
3. Wikipedia, s.v. “Flesch- Kincaid Readability Tests,” https://en.wikipedia.org/wiki/Flesch
%E2%80%93Kincaid_readability_tests (accessed February 14, 2016).
4. Flesch R, “How to Write in Plain English,” University of Canterbury, http://www.mang.canter-
bury.ac.nz/writing_guide/writing/flesch.shtml (accessed February 12, 2016).
5. Ibid.
6. In general, these ranges represent one standard deviation around the mean of our model revi-
sions. See our analysis in Chapter 7.
7. Bhambhani V, Muenke M, “Noonan Syndrome,” Am Fam Phys 89, no. 1 (January 2014): 40.
8. Carter R, “B Cells in Health and Disease,” Mayo Clinic Proc 81, no. 3 (March 2006), under “B Cells
and Disease States,” http://www.mayoclinicproceedings.org/article/S0025-6196(11)61466-3/
fulltext.
19
CHAPTER 1
Use normal sentence length
Modern English, especially written English, is full of bad habits which
spread by imitation and which can be avoided if one is willing to take the
necessary trouble.—George Orwell, Politics and the English Language1
What makes traditional medical writing hard to read? One thing is long sentences.
Controlling sentence length helps improve your reading ease score. Consider the
following excerpt from Watson and Crick’s 1953 article in Nature, which first
described the double-helix structure of DNA:
We wish to put forward a radically different structure for the salt of

deoxyribose nucleic acid. This structure has two helical chains each
coiled round the same axis (see diagram). We have made the usual chem-
ical assumptions, namely, that each chain consists of phosphate di-ester
groups joining β-d-deoxyribofuranose residues with 3’, 5’ linkages. The
two chains (but not their bases) are related by a dyad perpendicular to
the fibre axis. Both chains follow right-handed helices, but owing to the
dyad, the sequences of the atoms in the two chains run in opposite direc-
tions. Each chain loosely resembles Furberg’s model No. 1; that is, the
bases are on the inside of the helix and the phosphates on the outside.2
(wseg = 118/19.6/52.7/10.0)
Even though this excerpt describes a complex topic in molecular biology, the aver-
age sentence length is less than 20 words. The reading ease score is 52.7 and the
grade level is 10.0.
A. Keep sentence length 15 words average,

25 words maximum
Medical writing often needs to use long medical terms. Because of this, achiev-
ing a good reading ease score requires keeping the average sentence length under
19
20
control. We think this means about 15 words average and 25 words maximum. Of
course, when we say, 15 words average, we mean some sentences are longer and
some shorter.
There are exceptions. For example, you might write a longer sentence if your
sentence ends in a list, or if you quote the long title of a study. Generally, if a
sentence runs more than 25 words, it covers more than one idea and you should
consider splitting it into two or more sentences. You’ll find it’s often easy to split
up a long sentence to make shorter ones.
Writing style experts recommend mixing long and short sentences.3 Many
short sentences, one after another, can seem choppy and distract the reader.
Many long sentences, one after another, become tedious to read.
N O S C I E N T I F IC R E A S ON F OR L ON G SE N T E NC E S
The science of reading ease tells us long sentences tend to be hard to read. In this
respect, there is a good scientific reason to avoid long sentences. A complex idea
may take longer to explain, but sentences of moderate length will serve just as
well or better.
C OMMON W I SD OM AB OU T SE N T E N C E L E NGT H M AY NO T
A PPLY TO M E DIC A L W R I T IN G
Experts in other fields recommend different “ideal” average sentence lengths

ranging from 15 to 25 words.4 Some writing experts actually teach techniques
for writing long sentences.5 This may do for fiction, which seeks to entertain,
surprise and delight the readers and which tends to use fairly short words.
However, studies show that, of all kinds of writing, fiction tends to use the
shortest sentences.6
Trying to write longer sentences may work in other fields, but medical science
writing calls for keeping average sentence length at the shorter end of range of
“ideal” average sentence lengths.
There are several reasons why this is so. First, unlike fiction, medical writ-
ing seeks primarily to inform or persuade the reader and not to entertain.
Second, medical writing uses long terms to describe diseases, medicines,
chemicals and processes. These long terms tend to lower reading ease scores.
An author should compensate for these scientific terms by avoiding need-
lessly long sentences.
Third, unlike most other fields (e.g., law and government regulation) medi-
cal writing reaches a global audience. Many, if not most, readers of medical sci-
ence journals don’t speak English as their first language. An article published in
a major English-language journal will be read by doctors throughout the world.
Even many doctors and medical scientists who work in English-speaking coun-
tries originally came from somewhere else.
21
Use Normal Sentence Length 21
Because of their unique subject matter, terms and audience, a medical author
should strive for an average sentence length of about 15 words and a maximum
sentence length of about 25 words.
Exercise 1. A. Keep sentence length 15 words average, 25 words maximum

For each sentence:
1. Read it out loud and count the number of words.

2. Revise to make shorter sentences. Make any other changes you can think of to
3. Count the number of words for each new sentence and the total number of
words. Compute the average number of words per sentence.
1. But the physician’s subsequent choice to designate the hospi-

tal discharge as against medical advice and pursue the formalized
process associated with it (eg, specialized discharge forms) has no
evidence-based utility for patient care, is not legally required, and has
been shown to be associated with reduced willingness for the patient
to return for future care.7
2. Our Review will focus on advances in understanding of COPD and
its risk factors, prevalence, and natural history since these Reviews
were published, address some of the questions that still persist, and
raise some of the issues that health-care planners will have to con-
sider as the burden of COPD increases as the world’s population ages.8
3. Since CYP2C9 explained 6 to 10 percent of the variability in these
two patient samples, the VKORC1 genotype appears to be the most
important genetic factor determining variability in warfarin dose: in
both clinical populations its effect was approximately three times
that of the CYP2C9 genotype.9
4. A systematic review of randomized clinical trials of multiple risk
factor interventions for preventing ischaemic heart disease had a
modest effect on changes in lifestyle factors, cholesterol concentra-
tions, and blood pressure—the last two mainly owing to the drug
treatment used—but no significant effect on long term mortality
due to ischaemic heart disease.10
5. Over the past 100 years, the science of exercise has grown from
seminal discoveries documenting the effects of exercise intensity
on vascular control, heat production, oxygen requirement, and
lactic acid dynamics—which led to Nobel Prizes in physiology or
medicine in 1920 (August Krogh, Denmark) and 1922 (A.V. Hill,
United Kingdom and Otto Meyerhof, Germany)—to our modern-day
22
understanding that one’s cardiorespiratory fitness (indexed by one’s

maximum rate of oxygen consumption) is among the most powerful
predictors of morbidity and mortality.11
6. In a prospective study in the Netherlands that followed more than
30,000 students 10 to 14 years of age for up to three years, annual scoli-
osis screening in addition to the usual biennial health checkup detected
no cases of idiopathic scoliosis requiring surgery, and the authors con-
cluded that additional annual scoliosis screening was not needed.12
B. Keep the subject and verb close together in

the first seven or eight words
One thing that makes a sentence hard to read is having the subject and verb far
apart. (See the Glossary if you need a refresher on the terms subject and verb.)
This is especially true if the subject or verb is buried deep within a long sentence.
Putting the subject and verb close together, near the start of the sentence, is
a simple way to improve reading ease. This avoids the common problem of the
reader losing track of the subject or verb and having to go back.13,14
Exercise 1.B. Keep the subject and verb close together in the first seven or
eight words
For each sentence:
1. Read it out loud and underline the subject once and the verb twice. Count
the number of words that come between them. Do the subject and verb come
within the first eight words?
2. Revise to make new shorter sentences. Each new sentence should have its sub-
ject and verb close together within the first seven or eight words. Make any
other changes you can think of to improve reading ease.
1. Although SDM is well accepted in overtly value-laden clinical deci-

sions such as prostate-specific antigen testing and mammography
screening, the principles of SDM apply to a broad range of health care
decisions, discharges against medical advice included.15
Note: SDM stands for shared decision making.
2. The only comprehensive effort to date to estimate summary

measures of population health for the world, by cause, is the ongo-
ing Global Burden of Diseases, Injuries, and Risk Factors (GBD)
enterprise.16
23
3. Although drug effect is a complex phenotype that depends on

many factors, early and often dramatic examples involving succinyl-
choline and isoniazid facilitated acceptance of the fact that inheri-
tance can have an important influence on the effect of a drug.17
4. Transparent reporting of review methods and detailed reporting
of the clinical and methodological characteristics of the included
studies and their results are important to enable a reader to judge
the reliability of both the review and the individual studies and to
assess their relevance to clinical practice and the meaning of the
results reported in the review.18
5. Baseline demographic characteristics and the distribution of peri-
operative ARDS risk factors or modifiers among those who did and
did not develop postoperative ARDS (first procedure only) after bleo-
mycin therapy are presented in table 2.19
Note: ARDS stands for acute respiratory distress syndrome.
6. Because patients often are reluctant to discuss traumatic events

and may avoid treatment as a result, it is important to elicit patient
preferences for treatment interventions.20
C. Put the main point first and then give

commentary, detail or support
Sometimes, a sentence starts with a dependent clause that gives some commen-
tary, detail or support before stating the main idea. This works well for a short
sentence. For example:
As with many autoimmune diseases, celiac disease is about two to three

times more common in women.21 (wseg=17/17.0/50.2/10.4)
It also makes sense to state your premises before you state a conclusion.
But when a writer starts a long sentence with a long dependent clause, it can
tax the reader’s powers of concentration. For example, consider the sentence
below where the main point starts 18 words into the sentence (in italics). We can
improve reading ease by breaking up the long sentence, putting the main point
first, and minimizing other long words (Table 1-1).
Sometimes a long sentence contains several key points. In such a case, it helps
to split up a long sentence into shorter sentences, with each key point near the
start of its own sentence.
If you want to help a busy reader understand your ideas, put the main point
first, and then give commentary, detail or support.
24
Table 1-1. Put the main point first
Original Revised
Although reducing heart rate For a patient with sepsis, a slow heart rate
will decrease myocardial oxygen may cut cardiac output and tissue perfusion.
consumption and will improve diastolic For a patient without sepsis, slowing the
function and coronary perfusion, heart rate slows the heart’s oxygen use. It
for patients with sepsis, an inadequate also helps diastolic function and coronary
chronotropic response may potentially perfusion. (wseg = 38/12.6/55.9/8.6)
negatively affect cardiac output and tissue
perfusion.i (wseg = 34/34.0/0.0/24.3)
i
Morelli A, et al. “Effect of Heart Rate Control with Esmolol on Hemodynamic and Clinical
Outcomes in Patients with Septic Shock,” JAMA 310, no. 16 (2013): 1689.
Exercise 1.C. Put the main point first and then give commentary, detail or support
For each sentence:
1. Read it out loud and underline the main point or each key point.
2. Count the number of words that come before the main point or each key point.
3. Does the main point come near the start of the sentence? If not, revise to
break up any long sentence. Put the main point first or put each key point
near the start of its own sentence. Make any other changes you can think of to
1. Although Ontario surgeons receive a premium and bill accurately

when they operate urgently between 12 AM and 7 AM, the times and
length of the overnight procedure and also how fatigued the surgeon
truly was when starting elective cases the next day were unknown.22
2. First, estimates for 2000 of under-5 mortality, measured as the
probability of death between 0 years and 5 years of age (5q0), and
mortality as a young adult or middle-aged adult, measured as the
probability of death between 15 years and 60 years of age (45q15),23
were developed after review of available vital registration, sample
registration, and census data and the application of the synthetic
extinct generation and growth balance methods to correct for under-
registration of deaths.24
3. The trial showed that pharmacogenetic-guided initiation of warfa-
rin therapy resulted in a greater percentage of time in the therapeu-
tic range, fewer excessive INRs, a shorter median time to therapeutic
INR, and fewer dose adjustments.25
Note: INR stands for international normalized ratio.

25
4. The participation rate was lower than we expected when we did the
power calculations; however, taking into account the fact that more
people than expected had an increased risk and received counselling
and that not even a trend to a reduction in ischaemic heart disease
was observed, we doubt that a participation rate of 70% would have
made any difference.26
5. Recommended CRC screening strategies fall in 2 broad catego-
ries: stool tests that primarily detect cancer, which include detection
of occult blood or exfoliated DNA, and structural tests, such as flexible
sigmoidoscopy, colonoscopy, and computed tomographic colonography,
which are effective in detecting both cancer and premalignant lesions.27
Note: CRC stands for colorectal cancer.
6. According to the AUA, the presence of three or more red blood cells
on a single, properly collected, noncontaminated urinalysis without
evidence of infection is considered clinically significant microscopic
hematuria.28
Note: AUA stands for American Urological Association.
Conclusion
Use normal sentence length to make your writing more readable. There is no sci-
entific reason to write a long sentence. On the contrary, the science of reading
ease indicates you should keep your sentence length moderate. We suggest the
right length for medical writing is about 15 words average, with a maximum of
about 25 words. Keep the subject and verb together in the first seven or eight
words. Put the main point first.
In the next chapter, we consider word length and its effect on reading ease.
Notes
1. Orwell G, “Politics and the English Language,” Horizon (April 1946). Available online, http://
www.orwell.ru/library/essays/politics/english/e_polit/.
2. Watson, “Molecular Structure of Nucleic,” Nature (see Intro, n. 11).
3. See for example Wydick, Plain English for Lawyers, 36 (see Intro, n. 8).
4. See for example Kimble, Lifting the Fog of, 71 (see Preface, n. 5); Wydick, Plain English for Lawyers,
36 (see Intro, n. 8).
5. See for example Brooks L, Building Great Sentences: Exploring the Writer’s Craft, (Chantilly,
VA: The Great Courses, 2008).
6. Cutts M, Oxford Guide to Plain English, 3rd ed. (Oxford: Oxford University Press, 2009), 2.
7. Alfandre D, Henning-Schumann J, “What is Wrong with Discharges Against Medical Advice (and
How to Fix Them),” JAMA 310, no. 22 (2013): 2393.
8. Mannino D, Bust A, “Global Burden of COPD: Risk Factors, Prevalence, and Future Trends,”
Lancet 370 (September 2007): 765.
26
9. Rieder M, et al. “Effect of VKORC1 Haplotypes on Transcriptional Regulation and Warfarin

Dose,” N Engl J Med 352, no. 22 (2005): 2290.
10. Jorgensen T, et al. “Effect of Screening and Lifestyle Counselling on Incidence of Ischaemic
Heart Disease in General Population: Inter99 Randomized Trial,” BMJ 348 (2014), under
“Introduction,” http://www.bmj.com/content/348/bmj.g3617.
11. Bamman M, et al. “Exercise Biology and Medicine: Innovative Research to Improve Global
Health,” Mayo Clinic Proc 89, no. 2 (February 2014): 148.
12. Horne J, Flannery R, Usman S, “Adolescent Idiopathic Scoliosis: Diagnosis and Management,”
Am Fam Phys 89, no. 3 (February 2014): 194.
13. Greene, Writing Science in Plain, 18–21 (see Preface, n. 7).
14. Goldbart R, “Scientific Writing as an Art and as a Science,” Environmental Health (March
2001): 22.
15. Alfandre, “What is Wrong with,” 2393 (see chap. 1, n. 7).
16. Murray C, et. al. “Disability-adjusted Life Years (DALYs) for 291 Diseases and Injuries in 21
Regions, 1990–2010: A Systematic Analysis for the Global Burden of Disease Study 2010,”
Lancet 380 (December 2012): 2198.
17. Weinshilbourn R, “Inheritance and Drug Response,” N Engl J Med 348, no. 6 (February
2003): 259.
18. Mallett S, et al. “Systematic Reviews of Diagnostic Tests in Cancer: Review of Methods and
Reporting,” BMJ 333 (August 2006), under “Reporting of Primary Study Details,” http://www
.bmj.com/content/333/7565/413.
19. Aakre B, et al. “Postoperative Acute Respiratory Distress Syndrome in Patients with Previous
Exposure to Bleomycin,” Mayo Clinic Proc 89, no. 2 (2014): 184–186.
20. Warner C, et al. “Identifying and Managing Posttraumatic Stress Disorder,” Am Fam Phys 88,
no. 12 (2013): 831.
21. Pelkowski T, Viera A, “Celiac Disease: Diagnosis and Management,” Am Fam Phys 89, no. 2
(2014): 99.
22. Vinden C, et al. “Complications of Daytime Elective Laparoscopic Cholecystectomies

Performed by Surgeons Who Operated the Night Before,” JAMA 310, no. 17 (2013): 1840.
23. We found the term “(45q15),” which refers to the probability of death between 15 years and 60 years
of age, confusing.
24. Wang H, et al. “Age-
specific and Sex- specific Mortality in 187 Countries, 1970–
2010: A Systematic Analysis for the Global Burden of Disease Study 2010,” Lancet 380
(2012): 2072.
25. Zineh I, Pacanowski M, Woodcock J, “Pharmacogenetics and Coumarin Dosing—Recalibrating
Expectations,” N Engl J Med 369, no. 24 (2013): 2274.
26. Jorgensen, “Effect of Screening and,” under “Strengths and Limitations of Study,” (see chap. 1,
n. 10).
27. Bujanda L, et al. “Effect of Aspirin and Antiplatelet Drugs on the Outcome of the Fecal
Immunochemical Test,” Mayo Clinic Proc 88, no. 7 (2013): 683.
28. Sharp V, Barnes K, Erickson B, “Assessment of Asymptomatic Microscopic Hematuria in
Adults,” Am Fam Phys 88, no. 11 (2013): 748.
27
CHAPTER 2
Prefer the short word
If you want to influence your reader with your ideas, resist the tempta-
tion to use long Latin-or French-based words where shorter ones will
do. Your message will be clearer and have much more impact.—Anne
Greene, Writing Science in Plain English1
This chapter talks about how word length affects reading ease. If you want to write
about medicine clearly and concisely, prefer the short word. Never use a big word
just because you can. Use a big word when it helps you explain the science clearly
and concisely. Impress people with your good ideas, not your big words.
A. Keep essential scientific terms; minimize

other long words
What makes medical writing different from everyday speech? One difference is,
everyday speech uses mostly one-or two-syllable words. Medical writing uses
many words with three syllables or more.
We call long words that help you write science clearly and concisely, essential
scientific terms. Minimize other long words. (We call a word, a long word if it has
three syllables or more. But this doesn’t include a two-syllable word that becomes
a three-syllable word by adding a common ending, such as -ed, -es, or –ing. For
example: expand—expanded; transfuse—transfuses; report—reporting.)
Essential scientific terms include names of diseases, medicines, chemicals and
processes. They also include titles for studies and programs and other important
scientific terms. How do you decide if a scientific term is essential? Every author
needs to use their best judgment about this, but we think an essential scientific
term meets four tests:
1. No shorter word serves just as well,

2. You can’t paraphrase in a few short words,
27
28
3. Doctors and other medical scientists use the term consistently (i.e., exclu-
sively), and
4. It’s easy to look up in a standard reference.
Using these tests, atrial fibrillation qualifies as an essential scientific term since:
1. We can’t think of a shorter word that serves just as well,

2. We can’t paraphrase it in a few short words,
3. Doctors use it consistently, and
4. If you look up atrial fibrillation in a standard reference, you will find the right
meaning right away.
Using these same tests, pulmonary does not qualify as an essential scientific
term, since the word lung often serves just as well, and doctors use both terms.
Exercise 2.A. Keep essential scientific terms; minimize other long words

For each sentence:
1. Read it out loud and underline each long word (i.e., three syllables or more, not
counting common word endings).
2. Double underline each word you consider an essential scientific term.
3. For any word you underlined only once, try to substitute a shorter word or
paraphrase using a few shorter words. Make any other changes you can think
of to improve reading ease.
1. Therefore, it is not possible to determine whether all drug-eluting

stents could benefit from short-term regimens of dual antiplatelet
therapy based on the previous trials reported.2
2. A careful assessment of the demographic evidence on the levels of
age-specific mortality is an integral component of any Global Burden
of Disease Study: such analyses require the sum of deaths from spe-
cific causes to equal the independently assessed level of mortality
from all causes, for every age and sex group.3
3. Most of the pharmacogenetic traits that were first identified were
monogenic—that is, they involved only a single gene—and most were
due to genetic polymorphisms; in other words, the allele or alleles
responsible for the variation were relatively common.4
4. Diagnostic accuracy is essential for good therapeutic treatment.5
29
Prefer the Short Word 29
5. The 2 objectives of the study were (1) to examine the association
between physical activity and dietary behavior and (2) to exam-
ine the potential combined effect of physical activity and dietary
behavior on biological (eg, total cholesterol) and health (eg, waist
circumference) markers.6
6. Assuming vasal disruption and occlusion have been adequately
achieved during surgery, and assuming the patient adheres to using
another contraceptive method while awaiting confirmation of sterility,
true causes of vasectomy failure include recanalization (early and late)
and, more rarely, aberrant anatomy (e.g., the presence of a third vas).7
B. Cite a common medical term once

to avoid confusion
Many common medical terms don’t qualify as an essential scientific term since
they have a plain English equivalent or can be paraphrased in a few shorter words.
Think of urticara (hives), angioedema (giant hives), cardiovascular disease (heart
disease), cerebrovascular disease (blockage of arteries to the brain), and ventricular
hypertrophy (heart wall thickening).
There may be times when you hesitate to write in plain English, because you
feel it might confuse a reader if you don’t use a common medical term. For exam-
ple, consider the sentence: “The patient had tachycardia and hypertension, which
are signs of sympathochromaffin system discharge characteristic of pain.” None
of the words underlined qualifies as an essential scientific term, but you might
think a reader could be confused if you didn’t use the terms tachycardia, hyperten-
sion and sympathochromaffin system discharge.
Where this is the case, we suggest writing in plain English, but stating the
common medical term once in parentheses. For example: “The patient had a rapid
heart rate (tachycardia) and high blood pressure (hypertension), which are signs
of the ‘fight-or-flight response’ (sympathochromaffin system discharge) common
for patients in pain.” After you refer to the common medical term once, you are
free to continue writing in plain English. This approach saves time for all readers,
and helps reach the widest reasonable audience, but without unduly perpetuating
C. Write a compound word to promote reading ease

and show how you pronounce it
One type of long word that often adds to medicus incomprehensibilis is the com-
pound word. A compound is a word formed by combining two or more words, or a
30
word plus a prefix. Medical writing uses many compounds formed from Latin or
Greek root words. Try to handle compounds carefully, since:
• The longer the word, the lower the reading ease, and the less familiar it is
likely to be.
• Any gap between how a word is written and how it is pronounced may slow
down a reader.
• A reader may fail to see a word is a compound and puzzle over its meaning.
• An uncommon compound may be difficult to look up.
An occasional long compound may help explain the science clearly. But if you try
to use every big word you know, it will lead to a severe case of medicus incompre-
hensibilis. You will lose many readers and waste other readers’ time.
T H R E E M A I N T Y PE S OF C OMP OUN DS
There are three main types of compounds: the open compound, the closed compound,
and the hyphenated compound. With the open compound, words work together but
are written as separate words. Examples include: student nurse, 50 percent, and ref-
erence book. Since the words aren’t connected, you might not even think of them
as compounds.
With the closed compound, words, or a word plus a prefix, are written as one
word. Examples include: multicell, hyperadrenergic, vasomotor and sinoatrial. For
some closed compounds, the meaning and syllable stress change. (For example,
compare the compound, straightforward with the separate words, straight and
forward. The compound has a different syllable stress and carries a slightly dif-
ferent meaning.) There is a general trend towards combining words to make
new closed compounds (e.g., on-line becomes online).8 In medical writing, this
same trend can sometimes lead to ungainly long words that add to medicus
incomprehensibilis.
With a hyphenated compound, words, or a word plus a prefix, are written
together but separated by a hyphen. Examples are: pre-menstrual, cost-effective,
self-reported. With the hyphenated compound, the component words are pro-
nounced as separate words.
S TA N DA R D U S AG E F OR C OMP OUN DS
For best reading ease, medical writing should follow standard usage regarding
compounds and hyphenation. These rules are detailed and complex. For example,
The Chicago Manual of Style contains 10 pages on compound words and hyphen-
ation. Focusing on a few key guidelines can help a medical writer reduce medicus
incomprehensibilis:
31
• Use caution in writing a word as a closed compound, unless that form is widely
accepted and pronunciation and reading ease are not at stake.
• A well-placed hyphen can often make for easier reading.
• Words that can be misread should be hyphenated.
• Where no ambiguity would result, a hyphen is not mandatory. It may be better
to write the word as an open compound.9
S T R AT E G I E S F OR M A N AG IN G C OMPOUNDS
1. Consider paraphrasing

Often, you can paraphrase using shorter words.
2. Consider writing the word as an open compound

When you need to use a compound word, sometimes an open compound conveys
the meaning just as well as a closed compound. It may also better reflect how you
pronounce and improve clarity and reading ease. For example, the word, ultra
rapid is commonly written as a closed compound. But this tends to obscure how it
is pronounced. It is four syllables long, which tends to lower reading ease.
We would prefer to write the word as an open compound, ultra rapid. To us,
ultra rapid sounds like two words. We also don’t see any special need to emphasize
the connection between the parts. This way, it is easier to read and pronounce. The
reading ease score also improves.
3. Consider writing the word as a hyphenated compound

Sometimes you can make a closed compound clearer by writing it as a hyphen-
ated compound. Consider, the closed compound, hyperadrenergic. Since it is six
syllables long, it will lower the reading ease in any sentence where it appears. This
compound joins the words, hyper- and adrenergic, but you can’t find hyperadren-
ergic in Stedman’s Medical Dictionary. Instead you have to look up the two parts.
Adrenergic is itself a closed compound made up of Latin and Greek roots.10 We
can make things clearer by writing this word as a hyphenated compound, hyper-
adrenergic. (Because people think of hyper-as a prefix, we would not try to write
hyper adrenergic as an open compound.)
Or, consider another closed compound, psychomotor. Anyone not familiar with
this word might try to say, psy-CHOM-o-tor, as though it rhymed with barometer.
As a result, they also might miss that it is formed from the words psycho and
motor. The right way to pronounce this word is, PSYcho-MOtor. It sounds like two
words, with the stress at the beginning of each word. We think it would be clearer
to write it as psycho-motor. This helps close the gap between how we write and how
we speak. It should be easier to read and understand.
Table 2-1 shows examples of some other closed compounds we would prefer to
write as hyphenated compounds to help show meaning and pronunciation more
clearly.
32
Table 2-1. Hyphenate to show how you pronounce
Closed compound Hyphenated compound

postherpetic neuralgia post-herpetic neuralgia
multinational multi-national
antiviral anti-viral
nonfatal non-fatal
Each of these examples might confuse a reader on first reading, if only for an
instant. For example, somebody might easily misread the word postherpetic as
POS-therpetic. Writing this word as a hyphenated compound, post-herpetic helps
the reader better see the units of meaning.
Some of the same considerations apply to prefixes. Consider the words decon-
taminate or deactivate. We pronounce them as if they were written as D-contaminate
and D-activate. (We don’t say DECK-kon-ta-mi-nate or DEAK-ti-vate.) For an inter-
national medical audience, we think it would be better to write de-contaminate and
de-activate.
W HE N WO ULDN ’ T W E HYPHE N AT E?
It takes judgment to decide when to use a hyphen. We wouldn’t create a hyphen-

ated compound when an open compound conveys the meaning just as clearly.
We would hesitate to hyphenate an essential scientific term. We also wouldn’t
hyphenate when a closed compound meets three tests:
1. It is fairly short—usually three syllables or less,

2. Writing it as a closed compound is well-accepted in common usage, and
3. There is little chance a reader might mispronounce the word.
Guided by these principles, we would not hyphenate prefix, react, prototype, ultra-
sound or phenotype. There may be other cases where you choose not to hyphenate
a compound word. But you should always try to write a compound word as clearly
as you can to help reduce medicus incomprehensibilis.
Some journals may have a house style for writing compounds that differs from
the advice we give here. If you take care to write compounds clearly, but an editor
or reviewer insists on using their house style, by all means, defer to their judgment.
Exercise 2.C. Write a compound word to promote reading ease and show how
you pronounce it
While you’re in Stockholm to accept your Nobel Prize, you dine with other prize
winners who speak English, but not as well as you. One of them asks for your help
to pronounce the words underlined in the examples.
33
1. For each word underlined, either add a hyphen to help show the word’s mean-
ing or pronunciation, or tell why you think adding a hyphen wouldn’t be
helpful.
2. Revise to make any other changes you can think of to improve reading ease.
1. Predefining a threshold value for heart rate is difficult because it

must be individualized in the context of the patient’s overall hemody-
namic status and any preexisting comorbidities.11
2. For example, the results of the Burden of Lung Disease (BOLD)
study—a multinational investigation of the prevalence of COPD using
a standard methodology and reported in this issue of The Lancet, show
that one of the highest prevalences of COPD was recorded in South
Africa, a country that also has a high prevalence of tuberculosis.12
3. Although a recent study showed that genotype-guided dosing
led to superior control of anticoagulation, the finding was based
on a comparison with a nonrandomized, real-world parallel control
group.13
4. Flavonoids are naturally occurring bioactive compounds that rep-
resent a constituent of fruits and vegetables, beyond calorie and mac-
ronutrient content, that could potentially influence body weight.14
5. One of the identified causes of delay in ET is multimodal imaging.15
Note: ET stands for endovascular therapy.
6. The recommended modalities are photodynamic therapy, radiofre-

quency ablation, or endoscopic mucosal resection.16
D. Omit any unnecessary word ending

Languages sometimes change a word ending to show how the word functions in a
sentence. For example, in modern English, we use s to show plural, and ’s to show
possession or connection. We also have Latin-origin words that show possession
or connection using other endings: gene/genetic, molecule/molecular, practice/ prac-
tical, muscle/muscular.
Languages tend to lose endings over time. We may ask, is it clearer or more cor-
rect to say mathematical calculation or math calculation? In our view, both are clear
and correct, but math is more concise. It sounds less formal, but not distracting
or un-professional. Since using math helps improve reading ease, we would prefer
it. Respecting a colleague’s time, by writing as clearly and concisely as possible, is
always the most professional way to write.
34
Which endings do you need to convey your meaning clearly and concisely?
Your judgment may be as good as ours or better. Whenever you can reasonably
leave off a word ending, it helps improve reading ease.
Exercise 2.D. Omit any unnecessary word ending

For each sentence, we underlined a word ending that shows possession or connec-
tion (e.g., cell/cellular; gene/genetic).
1. Read the sentence out loud. Do you think it would work to delete the end-
ing and just use the root word? Or else, to use a simpler word? Why or
why not?
2. Revise the sentence, making any other changes you can think of to improve
reading ease.
1. Given that this was an observational study, unmeasured con-

founding or hidden bias might exist.17
2. However, poverty is regarded as a surrogate measure for many fac-
tors that subsequently increase the risk of COPD, such as poor nutri-
tional status, crowding, exposure to pollutants including high work
exposures and high smoking rates (in countries of low and middle
income), poor access to health care, and early respiratory infections.18
3. The concept of pharmacogenetics originated from the clini-
cal observation that there were patients with very high or very low
plasma or urinary drug concentrations, followed by the realization
that the biochemical traits leading to this variation were inherited.19
4. Benefit from combined mammography and breast physical exami-
nation screening was found in women aged 50–64, but not in women
aged 40–49.20
5. If familial cardiomyopathy is suggested on the basis of his-
tory, genetic testing and referral to a genetic counselor should be
considered.21
6. Dosing must be tailored to the patient’s symptoms and inflamma-
tory markers because up to 13% of patients required higher initial
doses.22
E. Avoid the noun string

One writing habit that often involves long words and low reading ease is the noun
string. A noun string is a group of nouns and their modifiers put into long strings.23
Sometimes, one noun modifies another noun. Here are a few examples: early
35
childhood thought disorder misdiagnosis, pre-adolescent hyperactivity diagnosis, and

medication maintenance level evaluation procedures.24 When a sentence uses more
than three nouns in a row, it becomes harder to read.25 The reader can’t see at first
glance which word in the string is important.
W HE N SHO ULD YOU U SE A N OUN S T R ING?
You may want to use a short, familiar noun string to help name a complex con-
cept in just a few words. Otherwise, avoid long strings that are not already well
known.26 Avoid creating a new noun string.27
Exercise 2.E. Avoid the noun string

For each sentence:
1. Read it out loud and underline each noun string.

2. Revise the sentence to eliminate the noun string. Make any other changes you
can think of to improve reading ease.
1. The study by Vinden et al provides direct evidence from patient
outcomes that operating the night before is not associated with
increased complications for elective laparoscopic cholecystectomies
performed the following day.28
2. Performance is assessed in terms of rigorous out-of-sample predic-
tive validity testing based on the root-mean-squared error of the log of
the age-specific death rates, the percentage of time that trend is accu-
rately predicted, and the coverage of the uncertainty intervals (UIs).29
3. By targeting VKOR, the post-translational modification of the
vitamin K-dependent blood-coagulation proteins is impaired.30
Note: VKOR stands for vitamin K epoxide reductase.
4. All participants with an unhealthy lifestyle had individually tailored

lifestyle counselling at all visits (at baseline and after one and three
years); those at high risk of ischaemic heart disease, according to pre-
defined criteria, were furthermore offered six sessions of group based
lifestyle counselling on smoking cessation, diet, and physical activity.31
5. As in the original CMS readmission model derivation, patients
who were discharged and then died before being rehospitalized were
not counted as “failures”; that is, they were included in the analysis
but not assigned a readmission event.32
36
6. Tick paralysis, which results from gravid female bites, is a toxin-
mediated ascending paralysis that generally resolves after tick
removal.33
F. Don’t be afraid to start a sentence

with and or but
Medical articles often start a sentence by using a long conjunction, such as fur-
thermore, nevertheless, or however, in order to continue an idea started in the
previous sentence. It would improve reading ease to substitute and, but, or some
other short conjunction, but you almost never see this. This complete aversion
to starting a sentence with and or but is unwarranted.34 Sometimes and or but
sounds fine; sometimes it doesn’t.
Exercise 2.F. Don’t be afraid to start a sentence with and or but

For each exerpt:
1. Read it out loud and underline the long conjunction that starts the last
sentence.
2. Try replacing that long conjunction with and, but, or some other short word,
and read it out loud again. How does it sound to you?
1. Because patients with sepsis have a wide range of sympathetic

activation and responsiveness, giving a fixed dose of a β-blocker
would probably be less effective and potentially harmful if given to all
patients. Furthermore, because adrenergic stress persists as long as
the external stress (eg, infection or injury), treatment was continued
for the entire intensive care unit stay.35
2. The period since 1970 has been characterized by substantial het-
erogeneity in mortality transitions. Life expectancy in Japanese
women in 2010 was 85.9 years, and is probably higher in 2012.
However, the gain in the past 40 years was only 11 years, compared
with total improvements of two to three times more in other parts of
Asia (eg, the Maldives), the Middle East (especially Oman), and Latin
American (eg, Bolivia, Peru, and Guatemala).36
3. Our analysis does not address the issue of whether a precise ini-
tial dose of warfarin translates into improved clinical end points,
such as a reduction in the time needed to achieve a stable therapeu-
tic INR, fewer INRs that are out of range, and a reduced incidence of
37
bleeding or thromboembolic events. However, our study lays impor-

tant groundwork for a prospective trial and suggests that such a trial
should be powered to detect the benefits of incorporating pharmaco-
genetic information into the dose algorithm for patients who require
high or low doses—the subgroups in our study for whom dose esti-
mates based on the pharmacogenetic algorithm differed significantly
from those based on the clinical algorithm.37
4. Studies are observational in nature, prone to various biases, and
report two linked measures summarizing the performance in partici-
pants with disease (sensitivity) and without (specificity). In addition,
there is more variation between studies in the methods, manufactur-
ers, procedures, and outcome measurement scales used to assess test
accuracy than in randomized controlled trials, which generally causes
marked heterogeneity in results.38
5. This addresses not only the challenge of selection but also several
other challenges (eg, availability and recognition). However, partici-
pants noted several liabilities with this approach, including lack of
subspecialty specificity, additional uncompensated effort or institu-
tional expense, and potential for overreliance on the system.39
6. This study found that older patients who had severe neurocardio-
genic syncope (average of seven syncopal episodes in the previous two
years and asystolic pauses averaging 11 seconds) had a decreased time
to first syncopal event after a pacemaker was implanted. However,
this study was a manufacturer-funded trial and did not report the
total syncopal burden or number of falls.40
G. Avoid using a high percentage of long words

Common English tends to use an occasional long word. Medical writing needs to
use essential scientific terms to help the reader understand the science. This prob-
ably means using more long words than common English. In both cases, common
English and medical writing, the occasional well-chosen long word is probably not
a problem. The reader might not even notice.41
By contrast, some medical articles use far too many non-essential long words.
A high percentage of long words is a symptom of severe medicus incomprehensibi-
lis. Good science, expressed using essential scientific terms, helps to inform and
persuade the reader. It may even impress them. Inflated diction that serves no
scientific purpose usually fails to impress.42
Exercise 2.G. Avoid using a high percentage of long words

1. Underline each long word in the excerpt. Double underline each essential
scientific term.
38
2. Compute long words as a percentage of total words. (# long words/270

words = ___ %)
3. Revise to minimize long words but keep essential scientific terms. Break up
any long sentence and make other changes to improve reading ease.
4. Compute a new percentage of long words for the revised excerpt. How big a
change did you achieve? Do you feel your revision improved reading ease?
In a majority of cases, metabolism that is mediated by cytochrome

P-450 represents a deactivation pathway. For some drugs, however,
oxidation leads to conversion of a prodrug into an active compound.
A prime example is codeine (metabolized by CYP2D6); other exam-
ples include clopidogrel (metabolized by CYP3A4), cyclophosphamide
(metabolized by CYP2B6) and tamoxifen (metabolized by CYP2D6).
The major pathway of codeine consists of glucuronidation and
N-demethylation, whereas the CYP2D6-mediated O-demethylation
to produce morphine is a minor reaction. Nevertheless, the latter is
a crucial step in bioactivation, since the affinity of codeine for the
µ-opioid receptor is only 1/200 to 1/3000 that of morphine. Previous
studies have shown that the effects of codeine—analgesic, respira-
tory, psychomotor, and miotic—are markedly attenuated in people
with poor metabolism of CYP2D6. On the other hand, people with
ultrarapid metabolism, such as the patient described by Gasche et al.
in this issue of the Journal, produce greater amounts of morphine
from codeine and therefore may experience exaggerated pharmaco-
logic effects in response to regular doses of codeine. Similar effects,
albeit less dramatic, have been described in patients with ultrarapid
metabolism of CYP2D6 in response to routine doses of hydrocodone
or oxycodone, which are other opioids requiring CYP2D6 mediated
activation. These reports clearly illustrate the effect of CYP2D6
genetic polymorphisms on the action of codeine, ranging from virtu-
ally no effect in patients with poor metabolism to severe toxic effects
in those with ultrarapid metabolism. To put these observations into
perspective, these extremes of response might be relevant for some
10 to 20 percent of whites who have phenotypes associated with
either poor metabolism or ultrarapid metabolism.43
Conclusion
The two easiest things you can do to make your writing more readable are to use
normal sentence length and prefer the short word. Just as there is no scientific
39
reason to write a long sentence, there is no scientific reason to use many non-
essential long words. On the contrary, the science of reading ease indicates you
should prefer the shortest word that does the job. When you can master these two
lessons, you will have gone a long way toward improving reading ease.
Notes
1. Greene, Writing Science in Plain, 30 (see Preface, n. 7).
2. Feres F, et al. “Three vs. Twelve Months of Dual Antiplatelet Therapy after Zotarolimus-eluting
Stents: The OPTIMIZE Randomized Trial,” JAMA 310, no. 23 (2013): 1517.
3. Wang, “Age-specific and Sex-specific Mortality,” 2071–2072 (see chap. 1, n. 24).
4. Weinshilbourn, “Inheritance and Drug Response,” 529 (see chap. 1, n. 17).
5. Mallett, “Systematic Reviews of Diagnostic,” under “Introduction,” (see chap. 1, n. 18).
6. Loprinzi P, Smit E, Mahoney S, “Physical Activity and Dietary Behavior in US Adults and Their
Combined Influence on Health,” Mayo Clinic Proc 89, no. 2 (February 2014): 190.
7. Rayala B, Viera A, “Common Questions about Vasectomy,” Am Fam Phys 88, no.
11(2013): 759.
8. Chicago Manual of Style, 15th ed., s.v. § 7.84.
9. Ibid. §§ 7.84–7.90.
10. Adren [L. ad, to + ren, kidney] + [G. Ergon, work]; Stedman’s Medical Dictionary, ed. 28, s.vv.
“Hyper,” “Adrenergic.”
11. Morelli, “Effect of Heart Rate,” 1689 (see chap. 1, Table 1-1).
12. Mannino, “Global Burden of COPD,” 767 (see chap. 1, n. 8).
13. Pirmohamed M, et al. “A Randomized Trial of Genotype-Guided Dosing of Warfarin,” N Engl J
Med 369, no. 24 (2013): 2295.
14. Bertoia M, et al. “Dietary Flavonoid Intake and Weight Maintenance: Three Prospective
Cohorts of 124,086 US Men and Women Followed for up to 24 Years,” BMJ 352, no. i17
(2016), under “Introduction,” http://dx.doi.org/10.1136/bmj.i17.
15. Singh B, et al. “Endovascular Therapy for Acute Ischemic Stroke: A Systematic Review and
Meta-Analysis,” Mayo Clinic Proc 88, no. 10 (2013): 1064.
16. Zimmerman T, “Common Questions about Barrett’s Esophagus,” Am Fam Phys 89, no. 2
(2014): 96.
17. Vigen R, et al. “Association of Testosterone Therapy with Mortality, Myocardial Infarction, and
Stroke in Men with Low Testosterone Levels,” JAMA 310, no. 17 (2013): 1834.
20. Miller A, et al. “Twenty-Five Year Follow-up for Cancer Incidence and Mortality of the
Canadian National Breast Screening Study: Randomized Screening Trials,” BMJ 348 (2014)
under “Introduction,” http://www.bmj.com/content/348/bmg.g366.
21. Dunlay S, Pereira N, Kushawaha S, “Contemporary Strategies in the Diagnosis and Management
of Heart Failure,” Mayo Clinic Proc 89, no. 5 (May 2014): 663.
22. Caylor T, Perkins A, “Recognition and Management of Polymyalgia Rheumatica and Giant Cell
Arteritis,” Am Fam Phys 88, no. 10 (November 2013): 678.
25. See for example Follett W, Modern American Usage, (1966), quoted in A Dictionary of Modern
Legal Usage, Garner B, (New York: Oxford University Press, 1987), s.v. “Noun Plague,” 380.
27. Williams, Style: Lessons in Clarity, 69–70 (see Preface, n. 8).
28. Zinner M, Fresichlag J, “Surgeons, Sleep and Patient Safety,” JAMA 310, no. 17 (2013): 1808.
40
9. Murray, “Disability-Adjusted Life Years,” 2200 (see chap. 1, n. 16).

2
30. Furie B, “Do Pharmacogenetics Have a Role in the Dosing of Vitamin K Antagonists?” N Eng J
Med 367, no. 24 (2013): 2345.
31. Jorgensen, “Effect of Screening and,” under “Intervention,” (see chap. 1, n. 10).
32. Hummel S, et al. “Impact of Prior Admissions on 30-day Readmission in Medicare Heart
Failure Inpatients,” Mayo Clinic Proc 89, no. 5 (May 2014): 624.
33. Juckett G, “Arthropod Bites,” Am Fam Phys 88, no. 12 (2013): 844.
34. A Dictionary of Modern, Garner, s.vv. “And,” “But” (see chap. 2, n. 25).
35. Pinsky M, “Is there a Role for β-blockage in Septic Shock?” JAMA 310, no. 16 (2013): 1677.
36. Wang, “Age-specific and Sex-specific Mortality,” under “Discussion” (see chap. 1, n. 24).
37. The International Warfarin Pharmacogenetics Consortium, “Estimation of the Warfarin Dose
with Clinical and Pharmacogenetic Data,” N Eng J Med 360, no. 8 (2009): 760.
38. Mallett, “Systematic Reviews of Diagnostic,” under “Introduction” (see chap. 1, n. 18).
39. Cook D, Sorensen K, Wilkinson J, “Value and Process of Curbside Consultations in Clinical
Practice: A Grounded Theory Study,” Mayo Clinic Proc 89, no. 5 (2014): 610.
40. Denay K, Johanson M, “Common Questions About Pacemakers,” Am Fam Phys 89, no. 4
(2014): 281.
41. Kimble, Lifting the Fog of Legalese, 164 (see Preface, n. 5).
42. Ibid., 72.
43. Caraco Y, “Genes and the Response to Drugs,” N Eng J Med 351, no. 27 (2004): 2868.
41
CHAPTER 3
Omit any needless word
The ability to simplify means to eliminate the unnecessary so that the
necessary may speak.—Hans Hoffman1
Plain English writing means avoiding or omitting needless words that contribute
to medicus incomprehensibilis. In this chapter, we give exercises for you to practice
spotting and eliminating needless words. We also point out a few common mark-
ers for needless words: of and that.
A. Spot and omit needless words

The next exercise asks you to practice spotting and omitting needless words.
Exercise 3.A. Spot and omit needless words

For each excerpt:
1. Read it out loud. Strike any word you think unnecessary and do minor re-
arranging or editing, as needed.
2. Count the number of words you struck and calculate the percent reduction.
(For example, if you strike 3 words from a 30-word sentence, the reduction is
3/30 = 10.0%.)
1. In our study, we hypothesized that a heart rate range between 80/
min to 94/min was a sufficient compromise between improving car-
diac performance and preserving systemic hemodynamics.2
2. Although the definition states that this effect is in response to
noxious particles or gases, such as those in tobacco smoke, there is
also some evidence that infections can have an important role in the
presence of chronic inflammation in the lung.3
41
42
3. In the multiple linear regression analysis adjusted for clinically

important covariates, four of the five common haplotypes were found to
be independently associated with the warfarin dose (P≤0.05) (Table 1).4
4. With the emergence of new direct acting antivirals, the treatment
paradigm for hepatitis C virus (HCV) infection is currently undergo-
ing its greatest change since the discovery of the virus 25 years ago.5
5. Both dietary and physical activity behavior are independent pre-
dictors of numerous health outcomes among adults.6
6. For historical reasons, the American Indian/Alaska Native popu-
lation is particularly at risk of health and health care disparities. We
examined national data to understand how American Indians/Alaska
Natives use the health care system. To visualize the comparison we
employed an “ecology of health care” model which uses a relative
box size to indicate differences between populations. We compared
American Indians/Alaska Natives with the remaining U.S. population
on self-rated poor health (see accompanying figure).
This analysis reveals, as expected, the American Indian/Alaska Native
population to be significantly more rural and impoverished than the
rest of the U.S. population. In addition, American Indians/Alaska
Natives rate their health as poorer, yet they access the health care
system less often than the rest of the U.S. population. When they do
access the health care system, they more often enter through emer-
gency departments. Despite poorer health of American Indians/
Alaska Natives, the rates of primary care visits and hospitalizations
are similar to the rest of the U.S. population.7
B. Omit the needless of

The word of is a marker for wordiness. If you search for the word of in your draft,
you can often find a way to omit a few needless words (either of or other words).
Test this tip for yourself in the following exercise.
Exercise 3.B. Omit the needless of

For each sentence:
1. Read it out loud and underline the word of wherever it appears.

2. Revise to eliminate of or other needless words you see. Make any other changes
you can think of to improve reading ease.
1. The objective of our study was to conduct a randomized, multi-

center clinical trial to assess the effect of CPAP treatment on blood
43
Omit Any Needless Word 43
pressure values and nocturnal blood pressure patterns of patients

with resistant hypertension and OSA.8
Note: CPAP stands for continuous positive airway pressure. OSA stands for obstruc-
tive sleep apnea.
2. Estimates of the number of deaths in children younger than

5 years from the UNPD are substantially higher than are our esti-
mations; for 2005–2010, their estimates are 8 million deaths higher
(1–6 million per year).9
Note: UNPD stands for United Nations Population Division.
3. Recurring themes in pharmacogenetics include the presence of

a few relatively common variant alleles of genes encoding proteins
important in drug response, a larger number of much less frequent
variant alleles, and striking differences in the types and frequencies
of alleles among different populations and ethnic groups. 10
4. The quality of the studies varied considerably; many studies were
old, and few of the published studies provided sufficient detail to rep-
licate the intervention used.11
5. At the time that the 2004 algorithm was published, there were 2
available rigorous evidence-base reviews of the treatment of RLS/
WED prepared under the auspices of the Standards of Practice
Committee of the American Academy of Sleep Medicine.12
Note: RLS/WED stands for restless legs syndrome/Willis-Ekbom disease.
6. Systemic symptoms (low-grade fever, fatigue, malaise, and weight

loss) occur in 30% to 50% of patients.13
C. Omit the needless that

Like of, the word that can also be a marker for wordiness. In some languages, you
must use the equivalent of that to start a dependent clause (e.g., daβ in German, que
in French or Spanish, что in Russian, or ka in Latvian). But in English, sometimes
you need it and sometimes you don’t. If you don’t need it, it’s best to leave it out.
Table 3-1 shows examples of short sentences that are clear without using that
to introduce a clause. In other cases, you may need that to avoid confusion. But,
where this is the case, consider whether the sentence might be too long and con-
tain too many ideas. Perhaps, rather than keeping that, it would be better to split
up the long sentence to make shorter sentences.
44
Table 3-1. You don’t always need that to start a clause
With that Without that

The intern suggested that Raul The intern suggested Raul needed a CAT
needed a CAT scan. scan.
The test confirmed that Natalia The test confirmed Natalia was pregnant.
was pregnant.
Exercise 3.C. Omit the needless that

For each excerpt:
1. Read it out loud and underline the word that each time it occurs. Then decide
whether you think it’s necessary.
2. Strike out any that or any other words you think unnecessary. (If you need to,
do other minor editing or re-arranging to make this work.)
3. Then revise to make any other changes you can think of to improve reading ease.
1. Prevalence studies estimate that 38,054 patients had a diagnosis

of a primary malignant brain tumor in the United States in 2010.14
2. Deaths assigned to causes that are not likely to underlie causes of
death have been reassigned with standardized algorithms.15
3. Other states have simplified or eliminated special prescribing
rules (such as those requiring the use of triplicate prescription pads)
that were designed to control and monitor prescribing but that had
the (presumably unintended) effect of discouraging all prescribing of
controlled substances.16
4. It has therefore been clear for some time that more effective and
tolerable treatment regiments for HCV are needed.17
Note: HCV stands for hepatitis C virus.
5. A more recent analysis by Goyal et al revealed that both admis-
sion and post-admission hyperglycemia (admission glucose level
≤ 3.8 mmo/L) could predict 30-day death rate in patients with AMI.18
Note: AMI stands for acute myocardial infarction.
6. Overt hyperthyroidism that is inadequately treated is associated

with an increased risk of adverse maternal and neonatal outcomes
(Table 4).19
45
Omit Any Needless Word 45
Conclusion
This chapter covered the tip, Omit any needless word. This seems obvious but we all
need a reminder from time to time.
In the last three chapters, we looked at tips for improving your reading ease
score. In the next part of the book, we shift our focus to the concept of vivid
language.
Notes
2. Morelli, “Effect of Heart Rate,” 1689 (see chap. 1, Table 1-1).
4. Rieder, “Effect of VKORC1 Haplotypes,” 2289 (see chap. 1, n. 9).
5. Feeney E, Chung R, “Antiviral Treatment of Hepatitis C,” BMJ 349 (2014), under “Introduction,”
http://www.bmj.com/content/349/bmj.g3308.
6. Loprinzi, “Physical Activity and Dietary,” 190 (see chap. 2, n. 6).
7. Elise A G, et al. “Ecology of Health Care: The Need to Address Low Utilization in American
Indians/Alaska Natives,” Am Fam Phys 79, no. 3 (2014): 217.
8. Martinez-Garcia M, et al. “Effect of CPAP on Blood Pressure in Patients with Obstructive Sleep
Apnea and Resistant Hypertension: The HIPARCO Randomized Clinical Trial,” JAMA 310,
no. 22 (2013): 2408.
9. Wang, “Age-specific and Sex-specific Mortality,” 2087 (see chap. 1, n. 24).
11. Jorgensen, “Effect of Screening and,” BMJ, under “Introduction” (see chap. 1, n. 10).
12. Silber M, et al. “Willis-Ekbom Disease Foundation Revised Consensus Statement on the
Management of Restless Legs Syndrome,” Mayo Clinic Proc 88, no. 9 (2013): 978.
13. Caylor, “Recognition and Management of,” 677 (see chap. 2, n. 22).
14. Omuro A, DeAngelis L, “Glioblastoma and Other Malignant Gliomas: A Clinical Review,”
JAMA 310, no. 17 (2013): 1842.
15. Murray, “Disability-Adjusted Life Years,” 2200 (see chap. 1, n. 16).
16. Quill T, Meier D, “The Big Chill: Inserting the DEA into End-of-Life Care,” N Eng J Med 354,
no. 1 (2006): 1–2.
17. Feeney, “Antiviral Treatment of Hepatitis,” under “Indirect Acting Antivirals—Interferon Alfa
and Ribavirin,” (see chap. 3, n. 5).
18. Yang S, et al. “Association of Dysglycemia and All-Cause Mortality Across the Spectrum of
Coronary Artery Disease,” Mayo Clinic Proc 88, no. 9 (2013): 938.
19. Carney L, Quinlan J, West J, “Thyroid Disease in Pregnancy,” Am Fam Phys 89, no. 4 (2014): 276.
46
47
concept
2
USE VIVID L ANGUAGE
Belief is nothing but a more vivid, lively, forcible, firm, steady con-
ception of an object, than what the imagination alone is ever able to
attain.—David Hume1
The second concept for writing clearly and concisely is, Use vivid language. Using
vivid language means using language that is clear, detailed, powerful, full of life, or
strikingly alive. Good writing is lively. We trace the word vivid to the Latin words
for alive, spirited or animated.2
Medical writing is naturally interesting since it involves human life. The word
biology means the “study of life.”3 A good medical science article should reflect
the natural vitality of the subject. It should propel the reader along so they learn
faster, better and easier. No medical writer should ever do anything to destroy
this natural vitality by trying to make their writing dull and boring.
This is not to say a medical article should be poetic, thrilling, or entertaining,
the way other types of writing sometimes are. After all, a doctor or other scientist
reads a journal to learn, not to be entertained. Still, an author should never seek
the opposite extreme by boring, confusing, or frustrating a reader with dull and
barely readable prose. Yet, sadly, some authors seem to do just that.
Up until now, our tips on improving reading ease haven’t considered scientific
content. From now on, they will deal more and more with scientific content, but
they may not always improve reading ease.
In the next three chapters, we give tips on making medical writing more vivid.
4. Prefer active voice
5. Prefer concrete language
6. Observe the 1066 principle
48
Notes
1. Hume D, “An Enquiry Concerning Human Understanding: Skeptical Solution of these
Doubts,” in The Harvard Classics, ed. Charles W. Eliot (New York: Collier) 37: 344.
2. The word vivid comes from Latin vividus (animated, spirited), from vivere (to live), akin
to vita (life), Ancient Greek βίος (bíos, life); Wiktionary, s.v. “Vivid,” (accessed January 4,
2015) https://en.wiktionary.org/wiki/vivid.
3. The term biology (Bio-+ -logy) is a modern term coined by taking its components from
Ancient Greek βίος (bíos, bio-, life) + -λογία (-logía, -logy, branch of study, to speak). This
term or analogous terms arose in different European languages around 1800. The word
βίολογία did not exist in Ancient Greek. Wiktionary, s.v. “Biology,” (accessed January 4,
2015) https://en.wiktionary.org/wiki/vivid.
49
CHAPTER 4
Prefer active voice
Too often, aspiring professionals think they join the club only when they
write in the club’s most complex technical language. It is an exclusion-
ary style that erodes the trust a civil society depends on, especially in
a world where information and expertise are now the means to power
and control. It is true some research can never be made clear to merely
intelligent lay readers—but less often than many researchers think.
—Joseph Williams1
Traditional medical writing over-uses passive voice. If you want to write vividly,
prefer writing in active voice. Use passive voice, sparingly, only when it helps you
write clearly and concisely.
A sentence in active voice reflects the way we talk to other people every day, so
a reader finds it easy to follow. It also sounds more direct and vigorous than one
written in passive voice. Using active voice as a habit makes for forceful writing.
This is true, not just with a writing concerned mainly with action, but for any kind
of writing. When you revise a sentence in active voice, it usually becomes shorter.2
What do the terms active voice and

passive voice mean?
In case you need it, here’s a quick refresher on active and passive voice. The term
voice describes whether the subject of the sentence is doing the action or receiving
the action. When the subject does the action, the verb is in active voice. When the
subject receives the action, the verb is in passive voice.3 Table 4-1 shows examples
of passive and active voice.
Some sentences are neither active nor passive, since the subject neither does,
nor receives any action. Such a sentence may describe a state of being or a state
of possession.
49
50
Table 4-1. Examples of passive and active voice
Passive Active
Patients and clinicians were recruited We recruited patients and clinicians from 7
from 7 clinical sites.i clinical sites.
The final analysis was performed We analyzed the data according to the
according to the intention-to-treat intention-to-treat principle after the
principle after the enrollment period enrollment period ended for the study.
ended for the study.ii
Studies of HIV in women were either A study of HIV in women often covered a
routinely undertaken within study group made up of only female sex
populations of female sex workers workers, or included a large number of them.
or included a substantial number of
them.iii
Kravitz R, et al. “Patient Engagement Programs for Recognition and Initial Treatment of
i
Depression in Primary Care: A Randomized Trial,” JAMA 310, no. 17 (2013): 1819.

Annane D, et al. “Effects of Fluid Resuscitation with Colloids vs. Crystalloids on Mortality in Critically
ii
Ill Patients Presenting with Hypovolemic Shock: The CRISTAL Randomized Trial,” JAMA 310, no. 17
(2013): 1812.
iii
Beyrer, “An Action Agenda for,” Lancet, under “Introduction” (see chap. 4, n. 6).
A. Identify active and passive voice

How do you identify a sentence written in passive voice? A passive sentence has
two basic features, although both may not appear in every passive sentence. The
first is a form of the verb to be (for example: is, are, was, were, has been, will have
been, would be, etc). The second is a past participle (generally, with an -ed ending).4
(See Glossary.)
The three examples in Table 4-1 each use a form of the verb to be and a past
participle: in the first example, were recruited. In the second, was performed. In the
third, were … undertaken.
Exercise 4.A. Identify active and passive voice

For each sentence:
1. Read it out loud and underline the subject once and the verb twice. Draw
braces around any {past participle}.
2. Is the sentence active, passive or neither?
1. For those patients randomized to CPAP treatment, optimal CPAP

pressure was titrated in the sleep laboratory on a second night by
51
Prefer Active Voice 51
an auto CPAP device (REMstar Pro M series with C-Flex, Philips

Respironics) within a period of less than 15 days after the diagnostic
study to obtain a fixed CPAP pressure value, according to a previous
validation by the Spanish Sleep Network.5
Note: CPAP stands for continuous positive airway pressure.
2. A meta-analysis of data from 14 countries reported that transgen-

der female sex workers had a higher burden of HIV (27%) than other
transgender women (15%), male (15%), and female sex workers (5%).6
3. Warfarin binds to albumin, and only about 3% is free and pharma-
cologically active.7
4. After entry, the 9.6 kb viral genome undergoes cytoplasmic trans-
lation into a single polypeptide, which is subsequently cleaved into 10
viral proteins—three structural and seven non-structural.8
5. This sex difference is not clearly understood.9
6. Ticagrelor is recommended for combination therapy with aspi-
rin in patients who have acute coronary syndrome (unstable angina,
non-ST elevation myocardial infarction, or ST elevation myocardial
infarction) to reduce death from cardiovascular causes.10
B. Revise passive into active voice

Revising passive voice into active voice helps improve your writing. Why?
I T S O UND S MOR E N ATUR A L
More than any other technique, using active voice and specifying who performs
an action changes the character of your writing.11 Active voice sounds direct and
vigorous. A reader understands a sentence in active voice more quickly and eas-
ily, since it reflects how the human mind naturally thinks and processes infor-
mation.12 For example, saying, “We recruited patients and clinicians,” sounds
stronger and more specific than “Patients and clinicians were recruited.”
I T S AV E S WOR DS
An active sentence often saves words; a passive sentence often uses more words.
A whole document that uses just passive sentences can be 30% longer than one
that just uses active sentences.13
I T AVOI D S T HE WHO-D ONE-I T M YS T E RY
In addition to being shorter and more direct, an active sentence forces you to
name the actor or causal agent of your narrative. With a passive sentence, the
52
actor tends to go unnamed.14 There is a deep human instinct to find cause-and-

effect relationships. We naturally ask, what causes something to happen? The liter-
ary genre of the who-done-it mystery plays upon this instinct. Passive voice often
obscures the actor; this can send the reader on a mental wild goose chase trying
to solve the mystery of who-done-it. Writing in active voice and naming the actor
avoids this problem.
Exercise 4.B. Revise passive into active voice

For each sentence:
1. Read it out loud and underline the subject once and the verb twice. Draw braces
around any {past participle}. Count the number of words in the sentence.
2. Is the sentence active, passive or neither?
3. Revise each passive sentence to put it in active voice, making any other changes
you can think of to improve reading ease. Count the number of words in your
revision.
1. Daily 24-hour urine collections for volume and urinary sodium
excretion were performed for 72 hours.15
2. COPD can be classified with respect to both phenotype and disease
severity.16
Note: COPD stands for chronic obstructive pulmonary disease.
3. Once a drug is administered, it is absorbed and distributed to its

site of action, where it interacts with targets (such as receptors and
enzymes), undergoes metabolism, and is then excreted.17
4. Researchers have found evidence for bias related to specific design
features of primary studies of diagnostic studies.18
5. The incidence of major injury in each of the cohorts was calculated
per 10,000 person-years.19
6. The exact pathophysiologic mechanism for scoliosis is unknown.20
C. When should you use passive voice?

Use passive voice sparingly and purposefully.21 Passive voice has proper uses.22 It
can help keep the same or similar subjects in a series of sentences, thus making
for better flow and easier reading. For example, in the following journal excerpt,
the last sentence is in passive voice. This helps keep the narrative focused on the
study population.
53
We conducted a nested case-control study within the Kaiser Permanente

Northern California (KPNC) integrated health-care system, which pro-
vides comprehensive inpatient and outpatient services for approxi-
mately 3.3 million members. The KPNC membership approximates the
underlying census race/ethnicity and socioeconomic distributions of
the Northern California region. Prescription drug benefits are utilized
by more than 90% of members.23 (wseg = 56/18.6/0.0/19.0)
Passive voice can help move a word to a strategic part of the sentence to give
emphasis or to connect to a word in the preceding sentence. For example,
Shakespeare uses this technique in the opening line of his play, Richard III: “Now
is the winter of our discontent, made glorious summer by this sun of York.”
Passive voice can also help where the action is important, but the agent is not.
For example, the following excerpt defines “vitamin B12 deficiency” without tell-
ing who came up with this definition.
Vitamin B12 deficiency was defined as the presence of 1 of the fol-

lowing: the first diagnostic code for vitamin B12 deficiency, using
International Classification of Diseases, Ninth Revision codes 281.0 (per-
nicious anemia), 281.1 (other vitamin B12 deficiency anemia), 266.2
(specified at KPNC as vitamin B12 deficiency), or specific text diagno-
ses of vitamin B12 deficiency in the problem list; an abnormally low
value of serum vitamin B12; or a new and at least 6-month supply of
injectable vitamin B12 supplements.24 (Italics in original.) (wseg = 80/
80.0/0.0/20.9)
In this regard, passive voice can sometimes save words, as in a list of procedures,
where one item is manipulated in several different ways.25
S UM M A RY
Use passive voice, sparingly, when you have a good reason. Otherwise, use active.
D. Minimize forms of to be and to have

If you want to write vividly, search for the verbs to be and to have and revise to
minimize them. Sometimes you need them, and sometimes you don’t. If you
don’t need them, try to avoid them as main verbs.26 (Other forms of the verb to be
include be, am, is, are, was, were, have been, will be, etc. Other forms of the verb to
have include have, had, have had, had had, will have, will have had, etc.)
To be and to have often appear in a sentence that lacks action. (In this case,
they are called intransitive verbs, since the subject of the sentence neither acts
54
Table 4-2. A vivid sentence describes action
No action (intransitive verb) Action (transitive verb)

Smoking status was {categorized} as We noted for each patient whether
current, past, or never smoked.i they: smoke now, used to smoke, or never
(wseg = 10/10.0/52.8/8.3) smoked. (wseg = 15/15.0/84.4/5.2)
Although it can be transient, older An older patient tends to get persistent
persons are more likely to have tinnitus more often, though it sometimes
persistent tinnitus.ii (wseg = 16/16.0/ lasts only a short while. (wseg = 18/18.0/
16.1/14.9) 56.9/9.7)
The illness episode was {classified} as We counted an illness episode as VCD, if
VCD if any test was positive.iii (wseg = the blood tested positive. (wseg = 12/12.0/
12/12.0/53.6/8.7) 60.7/7.7)
Note: VCD stands for virologically confirmed dengue.
i
Aakre, “Postoperative Acute Respiratory Distress,” Mayo Clinic Proc 183 (see chap. 1, n. 19).
ii
Yew K, “Diagnostic Approach to Patients with Tinnitus,” Am Fam Phys 89, no. 2 (2014): 106.
iii
Villar L, et al. “Efficacy of a Tetravalent Dengue,” N Eng J Med 372 (2015), under “Procedures,” http://
www.nemj.org/doi/full/10.1056/NEJMoa1411037?query=TOC.
upon anything else, nor receives action.) Instead, they describe a state of being or
a state of possession. By contrast, a sentence that describes action sounds more
vivid. Table 4-2 shows examples of sentences we revised in active voice to show
action.
For each example in Table 4-2, the sentence on the left contains a form of the
verb to be (was, be or are). It sounds less vigorous than the revision on the right,
which uses a verb in active voice to describe action.
You also need to use a form of the verb, to have to write in the perfect tense;
but the simple past tense, which does not use have, often serves just as well. See
Table 4-3.
Table 4-3. Revising to eliminate to have
Perfect tense using have Simple past tense without have

No recent studies have evaluated We found no recent study on the use of
methylphenidate in the treatment methylphenidate to treat fatigue in a general
of fatigue in general palliative care palliative care setting. (wseg = 19/19.0/45.0/11.6)
settings.i (wseg = 16/16.0/16.1/14.9)
Onishi E, Biagioli F, Safranek S, “Methylphenidate for Management of Fatigue in the Palliative Care
i
Setting,” Am Fam Phys 89, no. 2 (2014): 124.

55
S UM M A RY
If you want to write vividly, look for sentences that use a form of to be or to have,
and try to re-write them in active voice using a verb that shows action.
Exercise 4.D. Minimize forms of to be or to have

For each sentence:
1. Read it out loud. Underline each form of the verb to be or to have. Draw braces
around any {past participle}.
2. Revise to eliminate any form of to be or to have, and instead, use a verb in active
voice. Make any other changes you can think of to improve reading ease.
1. A linear regression was used to assess all trends over time.27
2. Such assessment is not a straightforward addition of reported
causes. Because there are likely to be many more data reported for
levels of all-cause mortality than there are for individual causes, the
independent assessment of age-specific mortality is crucial to con-
strain the often less robust estimates of cause-specific mortality
within each population group defined by age and sex.28
3. The current hospitalist-ambulist division of general medical care
has made important contributions to patient care, but it leaves much
to be desired, especially with regard to personalization and continuity
of care.29
4. Although they are part of a randomized trial, the participants rep-
resent a selected group of people who have chosen to participate and
who attended the follow-up.30
5. Inhaled corticosteroids (ICSs) have had a central role in the man-
agement of asthma, even before publication of the first Guidelines for
the Diagnosis and Management of Asthma in 1991.31
6. The cremasteric reflex, which is elicited by pinching the medial
thigh, causes elevation of the testicle.32
E. Identify nominalization

One of the over-used writing habits that make medical writing hard to read is
nominalization. Nominalization is the process of making an abstract noun out of
a verb or adjective. (What is an abstract noun? It is a noun that relates to the
world of ideas, often something we can’t touch or see. By contrast, a red rubber
ball is a real-world, concrete object, since you can touch and see it. We’ll talk more
56
Table 4-4. Examples of nominalization
Verb becomes an abstract noun Adjective becomes an abstract noun

isolate => isolation careless => carelessness
expose => exposure competent => competency
complicate => complication orderly => orderliness
investigate => investigation clean => cleanliness
approve => approval sufficient => sufficiency
limit => limitation effective => effectiveness
recommend => recommendation effective => efficacy
replace => replacement sensitive => sensitivity
inform => information normal => normalization
negate => negative negative => negativity
about abstract and concrete in the next chapter.) Table 4-4 shows examples of
nominalization.
A nominalized verb tends to make for a long, abstract word. Because the word
is long, it can drive down the reading ease score. Because the word is abstract, it
tends to make the writing less vivid.
W HE N SHO ULD YOU U SE A N OMIN A L IZAT ION?
There are a few situations where nominalization helps you present good science
clearly and concisely. The first is where the nominalization refers to a name, fixed
expression, or well-known subject. Examples:
• Few problems so divide Americans as abortion on demand.

• The Equal Rights Amendment surfaced as an issue in past elections.33
The second is where the nominalization serves as a short subject that refers to
a previous sentence, and promotes a smooth flow of logic. Table 4-5 gives a few
examples of this.
Table 4-5. Using nominalization as a short subject to promote the smooth

flow of logic
Instead of With nominalization
The hospital administrator accepted The hospital administrator accepted the
what the doctor requested. doctor’s request.
All these things they argue depend These arguments all depend upon an unproven
upon an unproven claim. claim.
57
S UM M A RY
Use nominalization when you have a good reason; otherwise, avoid it.
Exercise 4.E. Identify nominalization

For each sentence, read it out loud and underline each instance of nominalization.
1. To curb such empirical use, a report from the Infectious Diseases
Society of America (IDSA) is calling for steps to boost the develop-
ment of better diagnostic tests, to reduce regulatory hurdles for new
tests, and to improve clinical use of infectious disease diagnostics.34
2. In sub-Saharan Africa and southeast Asia, peer or community
counselling and condom distribution among female sex workers was
estimated to be cost effective, at US$86 per infection averted and $5
per DALY averted (all costs from here expressed in 2012 US$), and
was more cost-effective than school-based education, voluntary coun-
selling and testing, prevention of mother-to-child transmissions, and
STI treatment.35
Note: DALY stands for disability adjusted life year. STI stands for sexually transmit-
ted infection.
3. We hypothesized that the administration of fixed-duration anti-

biotic therapy (4 days) after source control would lead to equivalent
outcomes and a shorter duration of therapy as compared with the tra-
ditional strategy of administration of antibiotics until 2 days after
the resolution of the physiological abnormalities related to SIRS.36
Note: SIRS stands for systemic inflammatory response syndrome.
4. Over 93% of participants in the control arm aged 40-49 returned
their annual questionnaire, whereas compliance with annual breast
examination screening for those in the control arm aged 50-59 varied
between 89% (for screen 2) and 85% (for screen 5); only question-
naires were obtained for 3% to 7% of the women.37
5. For many of these reasons, evidence-based reviews generally make
authoritative statements on the degree of evidence in support of the
use of each medication for a defined disorder, but they are not always
conducive to the development of practical algorithms for the manage-
ment of disorders of varying severity and a lengthy natural history.38
6. If history or examination findings raise concern for intracranial
lesions, magnetic resonance imaging of the brain can be useful for
further evaluation, with particular scrutiny of the skull base.39
58
Table 4-6. Minimize nominalization to improve vividness
Nominalized verbs Revised to reduce nominalization

The factors that can affect patient outcome Many factors affect patient outcome,
include complete macro-and microscopic including:
tumor resection (R0-resection), the depth of 1. how much of the tumor is removed
tumor infiltration (T-category), the presence (R0-resection),
of lymph node metastasis (N-category), and 2. how deep the tumor infiltrates
the presence or absence of lymphatic vessel (T-category),
invasion (L VI).i (wseg= 39/39.0/6.7/22.0) 3. whether the cancer metastasizes to
the lymph nodes (N-category), or
4. whether it gets into the lymph vessels
(L VI). (wseg= 39/39.0/94.3/2.3)
i
rücher B, M Kitajima, and J Siewert, “Undervalued Criteria in the Evaluation of Multimodal
B
Trials for Upper GI Cancers,” Cancer Invest 32, no. 10 (2014), under “Multimodal Therapy,” http://
www.ncbi.nlm.nih.gov/pmc/articles/PMC4266078.
F. Convert nominalization into a verb in active voice

You can often improve reading ease and make your writing more vivid by avoid-
ing nominalization. Since nominalization is a grammatical form, unless it is
part of a name or a fixed expression, a nominalized term is rarely an essential
scientific term.
The example in Table 4-6 uses many nominalizations. None of these nominal-
izations is an essential scientific term, since we can replace each one with a verb in
root form, or by substituting a plain-English equivalent.
S U MM A RY
Avoid nominalization unless you can give a reason why it helps you write clearly
and concisely.
Exercise 4.F. Convert nominalization into a verb in active voice

For each sentence:
1. Read it out loud and underline each nominalization.

2. Revise to convert the nominalization to a verb in active voice. Make any other
changes you can think of to improve reading ease.
1. Data from each trial were entered on an intention-to-treat basis

according to the recommendations of the Cochrane Collaboration
59
and the Preferred Reporting Items for Systematic Reviews and Meta-
analyses (PRISMA) statement.40
2. Accurate estimation of the number of deaths in each age and sex
group in a country, region, or worldwide is a crucial starting point for
assessment of the global burden of disease.41
3. For some drugs, however, oxidation leads to conversion of a pro-
drug into an active compound.42
4. After five years of counselling a significant effect on lifestyle was
seen, with a substantial reduction in the prevalence of smoking,
improved dietary habits, sustained physical activity (among men),
and a decrease in binge drinking.43
5. Much evidence has been amassed in support of asthma treatment
with ICSs.44
Note: ICSs stands for inhaled corticosteroids.
6. In its 2011 recommendation statement, the U.S. Preventative

Services Task Force did not find sufficient evidence for or against
screening for bladder cancer in asymptomatic adults45
Conclusion
If you want to write vividly, prefer active voice. When you can, revise passive into
active, minimize to be and to have, and convert nominalization into a verb in
active voice.
In the next chapter, we discuss the difference between concrete and abstract.
Notes
2. Strunk W, White E B, The Elements of Style (New York: Macmillan, 1979), 18–19.
4. Plain Language Action and Information Network, Federal Plain Language Guidelines (March
2011, Revised May 2011), 20–21, available at www.plainlanguage.gov.
5. Martinez-Garcia, “Effect of CPAP on,” 2409 (see chap. 3, n. 8).
6. Beyrer C, et al., “An Action Agenda for HIV and Sex Workers,” Lancet 385, no. 9964 (January
2015), under “Introduction,” http://www.thelancet.com/journals/lancet/article/PIIS0140-
6736(14)60933-8/fulltext.
7. Furie, “Do Pharmacogenetics Have a,” 2345 (see chap. 2, n. 30).
8. Feeney, “Antiviral Treatment of Hepatitis,” under “HCV Life Cycle and Natural Course,” (see
chap. 3, n. 5).
9. Iyer V, Lim K, “Chronic Cough: An Update,” Mayo Clinic Proc 88, no. 10 (2013): 1116.
10. Tsang K, Hartmark-Hill J, “Ticagrelor (Brilinta) for Secondary Prevention of Thrombotic
Events Following Acute Coronary Syndrome,” Am Fam Phys 88, no. 12 (2013): 822.
60
11. Plain Language Action and Information Network, Federal Plain Language Guidelines, 20 (see
chap. 4, n. 4).
12. Office of Investor Education and Assistance, A Plain English Handbook (Washington DC: US
Securities and Exchange Commission, August 1998), 21.
14. Ibid.
15. Chen H, et al. “Low-Dose Dopamine or Low-Dose Nesiritide in Acute Heart Failure with
Renal Dysfunction: The ROSE Acute Heart Failure Randomized Trial,” JAMA 310, no. 23
(2013): 2534.
19. Lai M, et al. “Long-Term Use of Zolpidem Increases the Risk of Major Injury: A Population-
Based Cohort Study,” Mayo Clinic Proc 89, no. 5 (2014): 590.
20. Horne, “Adolescent Idiopathic Scoliosis: Diagnosis,” 193 (see chap. 1, n. 12).
21. Office of Investor Education and Assistance, A Plain English Handbook, 19 (see chap. 4, n. 12).
22. Wydick, Plain English for Lawyers, 31–32 (see Intro, n. 8).
23. Lam J, et al. “Proton Pump Inhibitor and Histamine 2 Receptor Antagonist Use and Vitamin B12
Deficiency,” JAMA 310, no. 22 (2013): 2436.
24. Ibid.
26. Ibid.
27. Kawwass J, et al. “Trends and Outcomes for Donors Oocyte Cycles in the United States, 2000–
2010,” JAMA 310, no. 22 (2013): 2427.
29. Goroll A, Hunt D, “Bridging the Hospitalist-Primary Care Divide through Collaborative Care,”
N Eng J Med 372, no. 4 (2015), http://www.nejm.org/doi/full/10.1056/NEJMp1411416.
30. Jorgensen, “Effect of Screening and,” under “Introduction,” (see chap. 1, n. 10).
31. Scanlon P, “Pneumonia Associated with Inhaled Corticosteroid Use in Chronic Obstructive
Pulmonary Disease: Another Perspective,” Mayo Clinic Proc 89, no. 2 (2014): 139.
32. Sharp V, Kieran K, Arlen A, “Testicular Torsion: Diagnosis, Evaluation and Management,” Am
Fam Phys 88, no. 12 (2013): 836.
33. Williams, Style: Lessons in Clarity, 48–49 (see Preface, n. 8).
34. Kuehn B, “IDSA: Better, Faster Diagnostics for Infectious Diseases Needed to Curb

Overtreatment, Antibiotic Resistance,” JAMA 310, no. 22 (2013): 2385.
35. Beyrer, “An Action Agenda for,” under “Costing of a New Response” (see chap. 4, n. 6).
36. Sawyer R, et al. “Trial of Short-Course Antimicrobial Therapy for Intraabdominal Infection,” N
Eng J Med 372, no. 21 (2015): 1997.
37. Miller, “Twenty Five Year Follow-up,” under “Methods” (see chap. 2, n. 20).
38. Silber, “Willis-Ekbom Disease Foundation,” 978 (see chap. 3, n. 12).
39. Malaty J, Malaty I, “Smell and Taste Disorders in Primary Care,” Am Fam Phys 88, no. 12
(2013): 854.
40. Udell J, et al. “Associated Between Influenza Vaccination and Cardiovascular Outcomes in
High-Risk Patients: A Meta-Analysis,” JAMA 310, no. 16 (2013): 1712.
42. Caraco, “Genes and the Response,” 2868 (see chap. 2, n. 43).
43. Jorgensen, “Effect of Screening and,” under “Introduction” (see chap. 1, n. 10).
44. Scanlon, “Pneumonia Associated with Inhaled,” 139 (see chap. 4, n. 31).
45. Sharp, “Assessment of Asymptomatic Microscopic,” 747 (see chap. 1, n. 28).
61
CHAPTER 5
Prefer concrete language
Few people have the imagination for reality.—Johann Wolfgang
von Goethe
The practice of medicine deals with both the real world and the world of abstract
ideas, but traditional medical writing tends to be overly abstract. This chapter
presents tips for making your writing more vivid by preferring concrete language.
Making writing more vivid often goes hand-in-hand with improving reading
ease. The first step is to learn to tell the difference between abstract and concrete
language.
A. Identify abstract and concrete subjects

What do we mean by abstract and concrete? A concrete object is something from
the real world, like a doctor, a patient or a test tube. A doctor performs many real-
world actions, such as looking in a patient’s ear or throat, listening to their heart,
or making an incision.
By contrast, an abstract idea is “a theoretical way of looking at things; some-
thing that exists only in idealized form.”1 Abstract ideas include concepts, theo-
ries, calculations and procedures. A medical diagnosis is abstract, since it involves
developing a theory of the nature and cause of a condition based on analyzing a
patient’s signs, symptoms and test results.
One of the best-known sentences in the English language is the Pledge of
Allegiance of the United States. “I pledge allegiance to the Flag of the United
States of America, and to the Republic for which it stands, one Nation under God,
indivisible, with liberty and justice for all.”
In this pledge, the Flag is a concrete, real-world object you can see and touch.
The sentence also deals with some abstract ideas: the United States of America, the
Republic, the Nation, indivisible, liberty and justice. These ideas are important, but
hard to visualize. The pledge presents the flag as a vivid, colorful symbol to repre-
sent these ideas.
61
62
It’s not always easy to make a clear distinction between abstract and concrete.
For example, if you turn on a water tap, you can see whether the water runs fast
or slowly; this is a real-world observation. If you measure the amount of water
that flows within a particular time, you can compute the rate of flow; this is an
abstract calculation. (You might say, merely observing that the water is running
fast or slowly involves abstract thinking).
Table 5-1 gives examples of medical terms that describe the real world and the
world of abstract ideas.
Table 5-1. Examples of real-world and abstract medical terms
Real world World of abstract ideas

heart cardiac
blood blood type
balance hydrostatic equilibrium
heartbeat, rhythm, stethoscope pulse, heart rate
pump, flow circulate, blood pressure, arrhythmia
vein, artery, arteriole circulation, circulatory system, coronary
output
scalpel, cut, incision surgical procedure
pill, tablet, shot, injection, IV bolus dose, dosage, dosing, dosing regimen
tumor, cancer cell lymphoma (diagnosis)
“zapping” a tumor treatment, protocol, therapy
die, death mortality
mouth, throat, stomach, gut, small digestion, digestive system
intestine, large intestine
lung, breath, breathing pulmonary, respiration, respiratory system
cough (symptom) cold (diagnosis)
lymph node lymphatic system
nerve, spinal cord nervous system
sperm, egg, ovary, womb, testicle, reproductive system, embryology, ovulation
embryo
vertebrae, spine, bone, skeleton skeletal system
warm, hot, thermometer fever, temperature
pee, urine urinary tract, urinary output
biopsy pathology, pathological analysis
computer, calculator, ruler statistical analysis, standard deviation,
mean, median, mode, average, error,
measurement
63
Prefer Concrete Language 63
S UM M A RY
It’s important to develop a good sense of whether a word or phrase sounds

abstract or concrete. If you favor concrete language, it will help make your writing
more vivid.
Exercise 5.A. Identify whether a subject is abstract or concrete

For each sentence:
1. Read the sentence out loud. Underline the subject.

2. Is the subject abstract or concrete?
1. Strengths of this study include the relatively large sample size,
the prospective assessment of leukocyte telomere length with blood
samples collected prior to HCT and the availability of detailed covari-
ate data known to influence transplant outcome.2
Note: HCT stands for hematopoietic cell transplant.
2. Fourth, individuals with similar smoking and exposure histories

can vary a great deal in the severity of their disease and response to
intervention.3
3. Vitamin K plays a single role in human biology—as a cofactor for
the synthesis of γ-carboxyglutamic acid.4
4. Previous research on systematic reviews of diagnostic tests noted
poor methods and reporting.5
5. We defined AMI by the presence of an increase and/or decrease
of cardiac biomarkers (preferably troponin) with at least 1 value
above the 99th percentile of the upper reference limit together
with evidence of myocardial ischemia with at least 1 of the follow-
ing: (1) symptoms of ischemia, (2) electrocardiographic changes
indicative of new ischemia (new ST-T changes or new left bundle
branch block), (3) pathological Q waves on the electrocardiogram,
and (4) imaging evidence of new loss of viable myocardium or new
regional wall motion abnormality.6
Note: AMI stands for acute myocardial infarction.
6. Documentation of the history, physical examination, diagnostic

study results, clinical impression, and diagnostic reasoning is vital
not only for medical care, but also for legal purposes.7
64
B. Revise abstract into concrete

With medical writing, sometimes you need to talk about a real-world object (e.g.,
a patient, a cancer cell, or a placenta). Sometimes you need to talk about a real-
world event (e.g., a patient dies, bleeds, vomits). Sometimes you need to talk
about an abstract idea (e.g., a cause and effect relationship, a diagnosis, a pro-
cess). No matter what you’re writing about, you often have a choice about how you
frame the discussion. Using concrete language tends to make your writing more
vivid and memorable. It also favors shorter words and better reading ease scores.
One strategy for writing clearly and concisely is to make a sentence’s subject a
concrete noun rather than an abstract noun. Let’s consider an example:
Because of the estrogen-mediated increase in thyroid-binding globulin,

the increased volume of the distribution of the thyroid hormone, and
the placental metabolism and transport of maternal thyroxine, there is
a 20% to 40% increase in the thyroid hormone requirement as early as
the fourth week of gestation.8 (wseg = 47/47.0/0.7/24.8)
This sentence uses an abstract subject, there, which lays buried deep within the
long sentence.
We can do a few things to simplify this sentence. We can use a concrete sub-
ject, woman. We can put her at the start of the sentence. We can split the sen-
tence into shorter pieces. “A woman needs 20% to 40% more thyroid hormone
by her 4th week of pregnancy. Why? Because her estrogen causes an increase in
thyroid-binding globulin. This thyroid hormone spreads over an increased volume
of distribution. The placenta metabolizes and uses some of the mother’s thyrox-
ine.” (wseg = 45/9.0/47.3/8.8). With these changes, the main subject and verb
make natural points of interest that make the science more vivid and improve
reading ease.
Exercise 5.B. Revise abstract into concrete

For each sentence:
1. Read it out loud and underline the subject.

2. Revise to use a concrete subject and replace other abstract language with
concrete language. Make any other changes you can think of to improve
reading ease.
1. During the past several decades, mean maternal age at delivery of
a first infant has increased steadily to 25.2 years in the United States
and 30 years in Germany and Britain in 2009.9
â•‡ 65
Prefer Concrete Languageâ•…â•… 65
2.â•‡ Fifth, the airflow limitation or obstruction that happens in COPD

is caused by a mixture of small airway disease, parenchymal destruc-
tion (emphysema), and, in many cases, increased airways responsive-
ness (asthma).10
3.â•‡ The goal of therapy is to keep the INR in the therapeutic range,
since patients with an INR that is subtherapeutic are at increased risk
for thrombosis and patients with an INR that is supratherapeutic are
at increased risk for bleeding.11
Note: INR stands for international normalized ratio.
4.â•‡ Ischaemic heart disease remains a leading cause of morbidity and

mortality worldwide.12
5.â•‡ Antihypertensive medication use was retrieved from the internal
pharmacy-â•‰dispensing records.13
6.â•‡ Over the past 75 years, the number of U.S. women receiving pre-
natal care has steadily increased.14
C.â•‡ Use nouns and verbs to carry the weight

of meaning
Write using nouns and verbs, not with adjectives and adverbs.—â•‰Strunk and White15
Adjectives and adverbs can add important detail to a sentence, but use them spar-
ingly. Studies show that articles in other fields tend to use between 11% and 18%
adjectives and adverbs.16 When your writing exceeds this “normal” range, con-
sider cutting adjectives and adverbs and revise to focus more on nouns and verbs.
This helps make your writing more clear and vivid.
What makes a sentence clear? One thing is, the reader can quickly grasp how
each word relates to the others. Consider a short sentence that uses only nouns
and verbs: “Jane threw Dick the ball.” (wseg = 5/â•‰5/â•‰100.0/â•‰0.0). Since this sentence
uses just three nouns and one verb, the reader easily sees how each word relates
to the others.
• Jane (the subject) did the throwing

• threw (the verb) is what Jane did with the ball
• the ball (the direct object) is what Jane threw to Dick
• Dick (the indirect object) is the one to whom Jane threw the ball
Adding a few short adjectives provides vivid detail but hardly changes the reading
ease score: “Jane threw Dick the red rubber ball.” (wseg = 7/â•‰7.0/â•‰100.0/â•‰0.6). As
we add more adjectives and adverbs, we start to make the sentence more subtle,
66
interesting and complex. But we also start to cloud the relationship between
words: “Pregnant Jane impulsively threw the half-inflated red rubber ball to an
equally astonished Dick.” (wseg = 14/14.0/35.5/11.7).
FAVOR R E A L-W OR LD A DJE CT I V E S OV E R AB S T R ACT ONE S
Not all adjectives and adverbs are created equal. The real-world adjectives red and
rubber are short and simple. They give vivid details about the physical properties
of the ball and may help the reader to visualize it. On the other hand, impulsively
and astonished are longer, abstract words. They represent Jane and Dick’s states of
mind as Jane throws the ball, which may be hard to visualize. It takes more mental
effort for a reader to process these abstract ideas.
Jane’s pregnancy adds a further complication. Of course, pregnancy is a real-
world condition. Depending on how far along she is, Jane’s pregnancy may seem
like more of a physical state—easy to visualize—or a state of mind. This sentence
also implies a question, Is Jane’s impulsive behavior related to her pregnancy?
A novelist may choose to write this way to make the story more interesting, but a
medical author usually just wants to explain the science clearly.
S U MM A RY
Adjectives and adverbs can add important detail, but they also complicate. It takes
skill to manage this tradeoff and decide: What information is important? and How
much can I pack into one sentence? Overusing adjectives and adverbs—especially
abstract ones—can sap vitality from a sentence. Cutting down on adjectives and
adverbs is one way to make your writing more vivid. Adjectives and adverbs that
relate to physical properties often help make your sentence more vivid; those that
relate to abstract ideas may have the opposite effect.
Exercise 5.C. Use nouns and verbs to carry the weight of meaning

For each sentence:
1. Read it out loud and underline each adjective and adverb. Count the number
of words you underlined. Compute adjectives and adverbs as a percentage of
total words.
2. Revise to reduce the number of adjectives and adverbs. Make any other
changes you can think of to improve reading ease. Compute a new percentage
of adjectives and adverbs.
1. But the physician’s subsequent choice to designate the hospital

discharge as against medical advice and pursue the formalized process
67
associated with it (eg, specialized discharge forms) has no evidence-

based utility for patient care, is not legally required, and has been
shown to be associated with a reduced willingness for the patient to
return for future care.17
2. We used these advances, and a further extension of the Brass
relational model life tables, to develop a time series of annual age-
specific mortality rates for 187 countries from 1970 to 2010, includ-
ing uncertainty.18
3. The bimodal distribution of plasma isoniazid concentrations in
subjects with genetically determined fast or slow rates of acetyla-
tion in one of those early studies strikingly illustrates the conse-
quences of inherited variations in this pathway for drug metabolism
(Fig. 2).19
4. There was evidence of bias when primary studies did not provide
an adequate description of either the diagnostic (index) test or the
patients, when different reference tests were used for positive and
negative index tests, or when a case-control design was used.20
5. The use of nonergot dopamine agonists has become widespread,
but increasing experience with these drugs has revealed treatment-
limiting adverse effects, including the development of augmentation
and impulse control disorders.21
6. Support groups could be helpful with diet maintenance.22
D. Write in the singular

One simple way to make your writing more vivid is to write in the singular.23
This can also help boost your reading ease score. One thing that makes writing
abstract and hard to visualize is a big number. It’s easy to imagine one thing. It’s
a little harder to think of two or three. It’s harder still to imagine a dozen or a
hundred.
You can easily imagine one puppy. It’s harder to imagine a mother dog nursing
a litter of puppies. You probably can’t imagine 101 Dalmatians, as in the children’s
novel by Dodie Smith.24 For us, as in the novel, 101 is a math calculation. It’s an
overwhelming number of dogs, not something you can easily imagine.
If you write in the singular, you won’t be able to use the plural to avoid the awk-
ward he or she, as the case may be. The alternative we use is the singular pronoun,
they, which has historical precedent and seems to be gaining ground.25
Let’s consider examples of how writing in the singular can help make your writ-
ing more vivid. See Table 5-2. For each example, we took a sentence with a plural
subject, revised it to make the subject singular, and then, made other changes to
68
Table 5-2. Revising to use a singular subject
Plural subject Singular subject Other changes

Patients with symptoms The patient with The patient with
suggestive of acute symptoms suggestive symptoms that suggest
myocardial infarction of acute myocardial acute myocardial
(AMI) account for infarction (AMI) accounts infarction (AMI) accounts
approximately 10% of all for approximately 10% of for 1 in 10 ER visits.
emergency department all emergency department (wseg = 17/17.0/40.2/
(ED) consultations.i (wseg (ED) consultations. (wseg 11.8)
= 19/19.0/0.0/19.1) = 19/19.0/0.0/18.7)
Patients want and A patient wants and A patient expects their
expect to have a well- expects to have a well- surgeon to be competent
rested and competent rested and competent and well-rested when they
surgeon performing their surgeon performing their operate on them.
operations.ii (wseg = 14/ operation. (wseg = 15/ (wseg = 16/16.0/58.4/9.0)
14.0/29.4/12.6) 15.0/33.6/12.2)
Reichlin T, et al. “One-Hour Rule-Out and Rule-In of Acute Myocardial Infarction Using High-Sensitivity
i
Cardiac Troponin T,” Arch Intern Med 172, no. 16 (2012), under “Conclusions,” http://archinte.jamanet-
work.com/article.aspx?articleid=1309579.
ii
Zinner, “Surgeons, Sleep and Patient,” 1808 (see chap. 2, n. 28).
S U MM A RY
Writing in the singular is an easy way to write more vividly and improve
reading ease.
Exercise 5.D. Write in the singular

For each sentence:
1. Read it out loud and underline each element that makes something plural.
2. Where you can, revise to put the sentence in the singular. Make any other
1. All patients had a telephone assessment of vital status and rehos-
pitalization at 60 and 180 days from randomization.26
2. All these hypotheses probably have elements of truth since COPD is
a classic gene-by-environment disease with various manifestations that
include increased airways reactivity, a characteristic response to infec-
tions, abnormal cellular repair, and development of complications or
comorbid disorders.27
69
3. After the intake of identical doses of a given agent, some patients
may have clinically significant adverse effects, whereas others may
have no therapeutic response.28
4. Participants were referred to their general practitioner for medi-
cal treatment, if relevant.29
5. All patients had angiographically defined CAD with at least 1 ves-
sel that met the American College of Cardiology/American Heart
Association (AHA/ACC) class I or II indications for PCI, and only
those who received implants with drug-eluting stents were consid-
ered eligible for the study.30
Note: CAD stands for coronary artery disease. PCI stands for percutaneous coronary
intervention.
6. Tinnitus occurs in most persons with normal hearing who are

exposed to silence.31
E. Talk in terms of one doctor treating one patient

The goal of medicine is one human life at a time—to keep one patient healthy—to
heal one patient—to keep one patient alive—to relieve one patient’s suffering. No
matter what they do now, every doctor has seen a patient in an exam room and
can imagine this scene vividly.
Regrettably, traditional medical writing often adopts a needlessly abstract
way of talking about multiple patients, patient populations, or a disease with
no patient at all. Even where a study deals with a group of patients or multiple
patient populations, you can often frame the discussion in terms of one doctor
treating one patient.
The fields of genetics, pharmacology, epidemiology, and statistics deal with
complex problems that affect millions of lives. These problems are solved through
abstract thought. Still, framing an issue in terms of how it affects one patient
often brings it into sharper focus.
Let’s consider some examples in Table 5-3. For each example, we start with
a sentence that deals with multiple patients or no patient. We underlined the
subject once and the verb twice. Then we revised to talk about one patient. Then
we made other changes to improve reading ease.
Exercise 5.E. Talk in terms of one doctor treating one patient

For each sentence:
1. Read it out loud.
2. Revise to talk about one doctor treating one patient. Make any other changes
you can think of to improve reading ease.
70
Table 5-3. Revising to talk about one patient
Multiple patients/no One patient Other changes

patient
All children were Each child was scheduled We scheduled each child to
scheduled for visits at for a visit at months 0, 6 get the vaccine at months
months 0, 6, and 12 for and 12 for vaccination and 0, 6 and 12 with a follow-
vaccination and at month at month 13 for a follow- up blood sample at month
13 for follow-up blood up blood sample. (wseg 13. (wseg = 22/22.0/
sampling.i (wseg = 24/24.0/62.6/10.4) 69.1/9.0)
= 22/22.0/57.6/10.6)
For example, several For example, several If a patient has a raised
studies have used studies have used common high-sensitivity C-reactive
common genetic variants genetic variants near the protein (CRP), does it
near the gene encoding gene encoding C-reactive affect their risk of heart
C-reactive protein (CRP) protein (CRP) to show disease? Many studies
to show that raised high- that, if a patient has a have looked at common
sensitivity CRP is unlikely raised high-sensitivity CRP, gene variants near the
to affect the risk of heart it is unlikely to affect their gene coding CRP, and it
disease causally.ii (wseg risk of heart disease. (wseg seems the answer is, no.
= 32/32.0/23.6/17.9) = 37/37.0/29.8/18.2) (wseg = 38/19.0/65.1/8.8)
i
Villar, “Efficacy of a Tetravalent,” under “Procedures,” (see chap. 4, Table 4-2).
ii
Frayling T, “Statins and Type 2 Diabetes: Genetic Studies on Target,” Lancet 385 (2015): 311.
1. In a cohort of patients with septic shock and high risk of mortal-
ity, our open-label use of esmolol after initial hemodynamic optimi-
zation resulted in maintenance of heart rate within the target range
of 80/min to 94/min.32
2. Use of lung function to characterize severity is, currently, the best
system available to clinicians, but it clearly falls well short of being
ideal.33
3. The response to many drugs in common use varies greatly among
patients.34
4. These agents seem to facilitate the use of shortened courses of
combination interferon-free therapy, which are associated with high
(>95%) sustained response rates and relatively few toxicities.35
5. Lactate levels have become a useful marker for tissue hypoperfu-
sion and may also serve as an end point for resuscitation in patients
with sepsis and septic shock.36
71
6. Women with hypothyroidism should be counseled about the

importance of achieving euthyroidism before conception because of
the risk of decreased fertility and miscarriage.37
Conclusion
This chapter gave tips on preferring concrete language. Change an abstract sub-
ject into a concrete subject. Use nouns and verbs to carry the weight of meaning,
instead of adjectives and adverbs. Write in the singular. Talk in terms of one doc-
tor treating one patient.
In the next chapter, we talk about how to choose the right word so your reader
can tell right away whether you are talking about something abstract or concrete.
Notes
1. Wiktionary, s.v. “Abstract,” https://en.wiktionary.org/wiki/abstract (accessed December
2, 2014).
2. Gadalla S, et al. “Association between Donor Leukocyte Telomere Length and Survival After
Unrelated Allogeneic Hematopoietic Cell Transplantation for Severe Aplastic Anemia,” JAMA
313, no. 6 (2015): 600.
6. Yang, “Association of Dysglycemia and,” 931 (see chap. 3, n. 18).
7. Kodner C, Wetherton A, “Diagnosis and Management of Physical Abuse in Children,” Am Fam
Phys 88, no. 10 (2013): 673.
8. Carney, “Thyroid Disease in Pregnancy,” 273 (see chap. 3, n. 19).
9. Kawwass, “Trends and Outcomes for,” 2427 (see chap. 4, n. 27).
12. Jorgensen, “Effect of Screening and,” under “Introduction” (see chap. 1, n. 10).
13. Sim J, et al. “Characteristics of Resistant Hypertension in a Large, Ethnically Diverse Hypertension
Population of an Integrated Health System,” Mayo Clinic Proc 88, no. 10 (2013): 1101.
14. Zolotor A, Carlough M, “Update on Prenatal Care,” Am Fam Phys 79, no. 3 (2014): 199.
15. Strunk, The Elements of Style, 71 (see chap. 4, n. 2).
16. Liberman M, “Stop Hating on Adjectives and Adverbs,” Slate.com, September 10, 2013, http://
www.slate.com/blogs/lexicon_valley/2013/09/10/adjectives_and_adverbs_mark_twain_
suggested_killing_them_but_counting_modifiers.html.
17. Alfandre, “What is Wrong with,” 2393 (see chap. 1, n. 7).
18. Wang, “Age-Specific and Sex-Specific Mortality,” 2072 (see chap. 1, n. 24).
21. Silber, “Willis Ekbom Disease Foundation,” 977 (see chap. 3, n. 12).
22. Pelkowski, “Celiac Disease: Diagnosis and,” 104 (see chap. 1, n. 21).
23. Bryan Garner, seminar in the early 1990s.
24. Smith D, The Hundred and One Dalmatians, (London: Heinemann, 1956).
25. Wikipedia, s.v. “Singular They,” https://en.wikipedia.org/wiki/Singular_they (accessed Feb
ruary 23, 2016).
72
6.
2 Chen, “Low-Dose Dopamine or,” 2534 (see chap. 4, n. 15).
29.
Jorgensen, “Effect of Screening and,” under “Intervention” (see chap. 1, n. 10).
30.
Yang, “Association of Dysglycemia and,” 931 (see chap. 3, n. 18).
31.
Yew, “Diagnostic Approach to Patients,” 106 (see chap. 4, Table 4-2).
32.
Morelli, “Effect of Heart Rate,” 1688 (see chap. 1, Table 1-1).
33.
Mannino, “Global Burden of COPD,” 766 (see chap. 1, n. 8).
34.
Caraco, “Genes and the Response,” 2867 (see chap. 2, n. 43).
35.
Feeney, “Antiviral Treatment of Hepatitis,” under “Abstract” (see chap. 3, n. 5).
36.
Andersen L, et al. “Etiology and Therapeutic Approach to Elevated Lactate Levels,” Mayo Clinic
Proc 88, no. 10 (2013): 1129.
7. Carney, “Thyroid Disease in Pregnancy,” 273 (see chap. 3, n. 19).
3
73
CHAPTER 6
Observe the 1066 principle
The truth is, many journal editors and senior scientists believe that
unclear scientific writing is a serious problem.—Anne Greene, Writing
Science in Plain English1
Introduction
In English, we tend to use short words to talk about the real world and longer
words, more sparingly, to talk about abstract ideas. We call this tendency the 1066
principle. It applies to all kinds of writing, including medical writing. Observing
the 1066 principle can help an author write more vividly.
How did English get to be this way? In 1066, William, Duke of Normandy,
invaded England together with thousands of French-speaking knights, soldiers,
clerks and clergy. These Norman invaders only partly replaced the existing English
aristocracy. As a result, over the next few centuries, people living in England used
Anglo-Saxon (Old English) and Norman French side by side. Eventually, French
died out as a spoken language in England, while Anglo-Saxon took on many
French words. Modern English emerged in the late 1400’s as the London dialect
became standard and the printing press came to England.2
The great sorting out

As the Normans brought French words into everyday use in England, there was
often an Anglo-Saxon word with the same meaning. In some cases, English kept
one word and dropped the other. In other cases, English kept both words, but gave
each a slightly different meaning.3 We generalize about this great sorting out in
Table 6-1.
Let’s look at examples of the 1066 principle at work in modern English. See
Table 6-2. For each pair of words with a similar meaning, we use the short word to
talk about the real world and the longer word to talk figuratively or about a more
abstract idea.
73
74
Table 6-1. Anglo-Saxon & Norman French words
Anglo-Saxon Norman French

basic meaning specialized meaning
shorter words longer words
literal figurative
real-world abstract
vivid logical
Table 6-2. Real world vs. abstract
General idea Real world Abstract

Put on top Heinz and Hilda put the The GM Canada logo superimposes
tablecloth on the table. a maple leaf over the GM logo.
Drink John went to the pub to drink Sofia goes to the library to imbibe
beer. knowledge.
What’s left? Nick ate half the pizza and put 80 divided by 7 equals 11 with a
the rest in the refrigerator. remainder of 3.
Subtract If I have four apples and I eat How much is four minus one? Four
one, how many do I have left? minus one equals three.
Watering Betsy watered her flowers. Manuel irrigated the fields.
WAT E R V S . I R R IG AT E
Looking at the last example in Table 6-2, you might ask, isn’t irrigating a field a
real-world activity? It does involve putting real water on real plants. But to irri-
gate a field, whether in ancient Egypt or modern China, an engineer must do sev-
eral things: determine the water needs for the crop; find a stable water source;
plan and build a system of canals, pipes, and pumps to move the water from the
source to the fields, etc. If we consider these steps, we see irrigating a field is not
just a real-world activity. It involves complex activities guided by abstract plan-
ning and analysis.
T H E 1 0 6 6 PR I N C IPL E A PPL IE S TO M AT H
Medical writing often deals with math or statistics. We tend to state a story
problem in real-world terms. For example, “If I have two mice in the cage, and
I buy two more mice, then how many mice do I have?” When we talk about math
in the abstract, we use longer French or Latin-origin words (e.g., plus, minus,
75
Observe the 1066 Principle 75
Table 6-3. Three ways to state a math problem
Story problem in real-world terms Abstract in words Abstract in symbols

A surgical team operates on Bob. If 12 minus 11 12 − 11 = ____
they start with 12 sponges before the equals ____.
surgery, but count only 11 after the
surgery, how many sponges did they
leave inside Bob?
If the field hospital has 12 tents to 103 divided by 103 ÷ 12 = ____
house 103 Ebola patients, how many 12 equals ____.
patients must share each tent?
equals, divide, remainder, quotient, mode, median, standard deviation, frequency,

distribution).
We can often state a math problem three ways: as a real-world story problem,
as an abstract math problem in words, or as an abstract math equation in sym-
bols. Stating the problem in symbols makes it less vivid, but easier to solve. See
Table 6-3.
A. Prefer the short word to describe the real world

We can observe the 1066 principle at work in medical writing. In Table 5-1, in
Chapter 5, we looked at examples of medical terms that deal with the real world
or the world of abstract ideas. The words in the Real world column tend to be
short—mostly one or two syllables.4 There are also some longer terms, but they
tend to be names or essential scientific terms: artery, arteriole, biopsy, injection,
ovary, stethoscope, testicle and thermometer.
The words in the World of abstract ideas column tend to be longer—three or
more syllables.5 Most are not essential scientific terms, since we can often find a
way to say something similar in shorter, more concrete sounding words (e.g., mor-
tality/death rate, circulation/blood flow, surgical procedure/operation, arrhythmia/
abnormal rhythm).
Exercise 6.A. Prefer the short word to describe the real world

For each sentence:
1. Read it out loud and underline any long word. Double underline any word you
consider an essential scientific term.
2. Does the sentence describe the real world? If so, revise to use short words to
replace any long word other than an essential scientific term. Make any other
76
1. Herpes Zoster (HV), caused by the reactivation of latent varicella-

zoster virus (VZV) manifests as an acute, painful vesicular rash and is
often accompanied by chronic pain or postherpetic neuralgia.6
2. Fifth, the airflow limitation or obstruction that happens in COPD
is caused by a mixture of small airway disease, parenchymal destruc-
tion (emphysema), and, in many cases, increased airways responsive-
ness (asthma).7
3. Maintenance of nocturnal euglycemia is extremely important and
is challenging, since most cases of severe hypoglycemia occur at night.8
4. Similarly, patients with diabetes had a lower risk of arterial throm-
bosis than those without diabetes.9
5. Immune responses are orchestrated by a complex, continually
evolving cooperative network of mobile cells and their products.10
6. The initial workup for urticaria and angioedema is a history and
physical examination to determine a possible etiology.11
B. Prefer ’s to show real-world possession

or connection
In English, we have a few different ways to show possession or connection. We
tend to use the ’s ending to show real-world possession or connection. We tend
to use of to show abstract possession or connection. We also sometimes use other
word endings, or no word ending at all. But open any medical journal today and
you rarely find an ’s. This is a symptom of medicus incomprehensibilis, which results
from overusing abstract language.
Let’s look at some examples of real-world and abstract possession or connec-
tion. See Table 6-4.
W HE N I S I T BE T T E R TO U SE OF?
We recommend using of sparingly. For example, when you’re talking about some-
thing abstract and it doesn’t sound right to use ’s, because it sounds too literal.
Sir Arthur Conan Doyle was an ophthalmologist who published his first
medical article, Gelsemium as a Poison in the British Medical Journal in 1879.12 Sir
Arthur is better known for creating the most famous fictional characters of all
time, Sherlock Holmes and Dr. Watson.
Sir Arthur’s most famous Sherlock Holmes novel is called, The Hound of the
Baskervilles. Why didn’t he call the novel, The Baskervilles’ Hound, using s’ to show
real-world possession or connection? Because one of the book’s mysteries is
77
Table 6-4. Real-world vs. abstract possession or connection
Example Abstract or real-world?

The doctor’s white coat hangs on the Real world
door.
The patient’s blood pressure was high. Real world
The nurse ignored the patient’s This implies the patient’s complaint is real,
complaint of pain. but their pain might not be.
The nurse ignored the patient’s pain This implies the patient’s complaint and
complaint. pain are real.
The objectives of the current study An objective is an abstract idea. It is
were to identify distinct sets of appropriate to say, the objectives of the
functional trajectories in the year study, but we could also say, the current
immediately before and after a study’s objectives.
serious fall injury, to evaluate the A type of injury is an abstract idea. It would
relationship between the prefall and sound awkward to say, the injury’s type.
postfall trajectories, and to determine
whether these results differed based
on the type of injury, namely hip
fracture vs. other serious fall injuries.i
i
Gill T, et al. “The Course of Disability Before and After a Serious Fall Injury,” JAMA Intern Med 173,
no. 19 (2013): 1781.
whether or not the hound really exists. The title, The Hound of the Baskervilles,
leaves open the possibility the hound might just be a legend.
As a medical writer, you rarely want to sound vague or tantalize your reader
with a mystery. Instead, you want to present your ideas clearly and directly.
Favoring ’s to show real-world possession or connection is one useful tool.
Exercise 6.B. Prefer ’s to show real-world possession or connection

For each sentence:
1. Read it out loud and underline each word or word ending that shows posses-
sion or connection. Tell whether that possession or connection relates to the
real world or the world of abstract ideas.
2. Revise to drop any unnecessary word ending. Show real-world possession
using ’s. Make any other changes you can think of to improve reading ease.
1. Within each of these 2 cohorts, we compared the effectiveness of

each intervention with a control (sleep hygiene informational video).13
78
2. In this report and the accompanying appendix, we present the

data, methods and key findings of the Global Burden of Disease Study
2010 on levels, trends, and age patterns of mortality worldwide.14
3. Importantly, these trials all examine the initiation of therapy with
vitamin K antagonists and use as a primary end point the percentage
of time that a patient is within the therapeutic range during the ini-
tial phase of treatment.15
4. Systematic reviews of diagnostic studies involve additional chal-
lenges to those of therapeutic studies.16
5. Randomized clinical trials are essential to evaluate therapies that
reduce rather than eliminate a complication of a disease.17
6. The U.S. Preventive Services Task Force recommends routine HIV
screening, known as opt-out screening, regardless of patient or phy-
sician perception of risk for all persons 15 to 65 years of age, unless
a patient refuses.18
C. Use terms consistently; avoid elegant variation

For medical writing, precise meaning and clear understanding are vital. You may
have learned in school to vary terms to make your writing more interesting. This
is called, elegant variation.19 Elegant variation is used when reading ease is not a
concern and there is no possibility of confusion. For example, if you write a report
about Thomas Edison, you might refer to him as the Wizard of Menlo Park, so you
don’t keep saying, Edison did this, and Edison did that.
As with other types of technical writing, there is little place for elegant varia-
tion in medical writing. You will potentially confuse your reader if you use differ-
ent terms for the same concept. For example, if you use the term senior citizens to
refer to a group, continue to use this term throughout your article. Don’t substi-
tute another term, such as the elderly or the aged. Using a different term may cause
the reader to wonder if you are referring to the same group.
Don’t feel you need to use synonyms to make your writing more interesting.
While using different words may make writing more interesting, it may decrease
clarity.20
Elegant variation often violates the 1066 principle by pairing a concrete-
sounding word with an abstract-sounding word, or by using words with different
levels of abstraction. Either situation can confuse a reader.
Exercise 6.C. Use terms consistently; avoid elegant variation

For each sentence:
1. Read it out loud and underline any place in the text that talks about a similar
idea using different terms.
2. Revise to use consistent terms. Make any other changes you can think of to
79
1. Portenoy states the problem is a lack of studies, not positive

results. Even though there is minimal literature on long-term efficacy
of opioids for chronic noncancer pain, the few studies that have been
published have failed to find good evidence for efficacy.21
2. Estimates from WHO’s Global Burden of Disease and Risk Factors
project show that in 2001, COPD was the fifth leading cause of death
in high-income countries, accounting for 3.8% of total deaths, and
it was the sixth leading cause of death in nations of low and middle
income, accounting for 4.9% of total deaths.22
3. Many legislatures and regulatory boards have adopted model pain
statutes that encourage compliance with established standards for
prescribing of pharmacologic agents for pain and other symptoms
and that protect physicians who observe these guidelines from regu-
latory intrusion and possible prosecution.23
4. With the emergence of new direct acting antivirals, the treat-
ment paradigm for hepatitis C virus (HCV) infection is currently
undergoing its greatest change since the discovery of the virus
25 years ago. New data are routinely released for different combina-
tions of these new agents, each reporting exceptionally high sus-
tained response rates for an infection that was once notoriously
difficult to treat. It is therefore difficult (even for those practicing
in the field) to keep abreast of present treatment options, or what is
likely to be available in the next 12–18 months. Because newer anti-
virals have recently been licensed in the United States and Europe,
and the results of several promising large phase III studies have
been recently published, now is an opportune time to review the
current treatment landscape for HCV, and to anticipate how that
landscape might look in coming years.24
5. Men taking zolpidem are at an increased risk of major injury.
Compared with the corresponding comparison cohort, the male zol-
pidem user cohort had a higher risk of major injury.25
6. Celiac disease occurs in persons of European ancestry and in
those of Middle Eastern, Indian, South American, and North African
descent. It is rare in persons of Asian descent.26
D. Avoid using a long, Latin word to describe

the real world
Traditional medical writing sometimes ignores the 1066 principle by (1) using a
long word that sounds abstract to talk about the real world, or (2) using a short
80
word that sounds concrete to talk about something abstract. Either one can con-
fuse a reader by giving a false signal of abstract or real world.
FA L SE -SIG NA L WOR DS
Let’s consider some examples of false-signal words. Traditional medical writ-

ing uses the terms mediate, modulate and regulate to describe real-world, natural
processes. But, in common usage, these same words are mostly used to describe
a complex activity guided by human thought and analysis. (For example, “the
Securities and Exchange Commission regulates the stock market.”)
Consider the example in Table 6-5, which uses the terms mediated and modu-
lated. The false-signal words mediate and modulate help make this sentence seem
“abstract” or “theoretical.” (But nothing is more real-world than the way Fentanyl
acts in a patient’s body.)
The mixed message about whether Fentanyl mediates or modulates the patient’s
response to pain adds to the problem. By revising to eliminate the false-signal
words mediate and modulate, we can give this sentence a more “real-world” feel.
This makes the discussion clearer and easier to follow.
Social psychologists also use the terms mediator and moderator to describe real-
world factors that affect a person’s behavior. A 1986 journal article notes that
psychology researchers tend to get these terms mixed up. It then goes on for eight
pages explaining their proper use.27
Table 6-6 lists terms to describe when a person can perceive a stimulus. These
long words look abstract, but each talks about a real-world natural process. As
Table 6-6 shows, we can easily paraphrase in plain English.
Often, you can replace a word that sounds abstract with other words that
sound more concrete. If you need ideas on how to do this, try looking up forms of
mediate, modulate and regulate in Stedman’s Medical Dictionary.28
Table 6-5. Revising to minimize false signal words
Original Revised
Fentanyl-mediated or modulated responses Fentanyl acts on the μ-opioid receptor
involve action at the μ-opioid receptor as as an agonist at the dorsal horn. This
an agonist at the dorsal horn inhibiting action inhibits ascending pain pathways
ascending pain pathways in the rostral in the rostral ventral medulla. The
ventral medulla, increasing pain threshold, result is to increase the patient’s pain
and producing both analgesic and sedative threshold, reduce their pain, and calm
effects.i (wseg = 36/36.0/0.0/23.6) them. (wseg = 39/13.0 /56.9/8.5)
Ruan X, Chiravuri S, Kaye A, “Toxicological Testing when Evaluating Cases of Suspected of Acute
i
Fentanyl Toxicity,” Forensic Sci Med Pathol (July 7, 2016).

81
Table 6-6. Sense of perception
Long Latin terms Plain English (sense of____)

ophthalmoception sight
audioception hearing
gustaoception taste
olfacoception or olfacception smell
tactioception touch
nociception pain
equilibrioception balance
thermoception warmth
proprioception one’s own movement
S UM M A RY
English tends to use long Latin words, sparingly, to talk about abstract ideas.
Because of this, long words tend to signal abstract. Using a long word to talk about
the real world tends to confuse by sending a false signal.
If you want to write clearly, try to avoid false signal words. This helps make
your writing more vivid and easier for a reader to follow. It also helps break down
the walls that hamper the free flow of ideas between fields.
Exercise 6.D. Avoid using a long, Latin word to describe the real world

For each sentence:
1. Read it out loud. Underline the words mediate, modulate or regulate whenever
they appear in any form (e.g., regulatory).
2. Tell whether you think the sentence describes the real world or an abstract
idea. If you think the sentence describes the real world, revise to use shorter
or more concrete-sounding words. Make any other changes you can think of to
1. The risk of HZ is elevated by 1.5 to 2 times in patients with rheu-

matic and immune-mediated diseases such as rheumatoid arthritis
(RA) and Crohn’s disease.29
Note: HZ stands for Herpes Zoster.
2. We hypothesise that rare ADRB2 variants modulate therapeutic

responses to LABA therapy and contribute to rare, severe adverse
events.30
82
Note: ADRB2 stands for the β2 adrenergic receptor gene. LABA stands for long acting
β antagonist.
3. Among the 54 patients, we identified 5 who could willfully modu-

late their brain activity (Figure 1).31
4. Transparency of the regulatory system is also required to over-
come several dysfunctions in the drug industry’s behaviour.32
5. Some experimental support exists for the concept that the abil-
ity to discriminate between “self” and “nonself” involves learning to
respond aggressively when there are signals that suggest the pres-
ence of invasive pathogens and having effective regulatory mecha-
nisms for suppressing inflammatory responses when such signals are
absent.33
6. Urticaria and angioedema are thought to have similar underlying
pathophysiological mechanisms, with histamine and other mediators
being released from mast cells and basophils.34
Conclusion
Remember the 1066 principle. If you want to make your writing easier to under-
stand, talk about the real world using short words. Use long words, sparingly, to
talk about abstract ideas. Use ’s to show real-world possession or connection or
to show abstract possession or connection if it sounds okay. Avoid the elegant
variation.
Notes
1. Greene, Writing Science in Plain, 1 (see Preface, n. 7).
2. Baugh A, Cable T, A History of the English Language, 5th ed. (London: Prentice-Hall, 2002),
67–115.
3. Thus, e.g., in modern English, we use Anglo-Saxon-origin words, cow, calf, pig and sheep, to talk
about the animal, and French-origin words, beef, veal, pork, and mutton, to talk about the food.
4. Some are Anglo-Saxon origin: heart, heartbeat, blood, flow, beat, pill, shot, dies, death, gut, lung,
cough, egg, and womb. Some are short French or Latin-origin: vein, IV bolus, tumor, cancer,
spinal cord, urine and sample.
5. Most are French or Latin origin; only a few are Anglo-Saxon.
6. Zhang J, et al. “Association between Vaccination for Herpes Zoster and Risk of Herpes Zoster
Infection among Older Patients with Selected Immune-Mediated Diseases,” JAMA 308, no. 1
(2012): 43.
8. Phillip M, et al. “Nocturnal Glucose Control with an Artificial Pancreas at Diabetes Camp,” N Eng
J Med 368 (2013): 825.
83
9. Donzé J, et al. “Impact of Sepsis on Risk of Postoperative Arterial and Venous Thromboses: Large
Prospective Cohort Study,” BMJ 349 (2014), under “Subgroup Analysis for Arterial
Thrombosis,” http://www.bmj.com/content/349/bmj.g5334.
10. Carter, “B Cells in Health,” under “Article Outline,” (see Concept 1, n. 8).
11. Schaefer P, “Urticaria: Evaluation and Treatment,” Am Fam Phys 83, no. 9 (2011), under
“Evaluation,” http://www.aafp.org/afp/2011/0501/p1078.html.
12. Wikipedia, s.v. “Arthur Conan Doyle,” https://en.wikipedia.org/wiki/Arthur_Conan_Doyle
(accessed June 15, 2015).
13. Kravitz, “Patient Engagement Programs for,” 1819 (see chap. 4, Table 4–1).
17. Sniderman A, et al. “The Necessity for Clinical Reasoning in the Era of Evidence-Based
Medicine,” Mayo Clinic Proc 88, no. 10 (2013): 1108.
18. Sherin K, et al. “What is New in HIV Infection?” Am Fam Phys 89, no. 4 (2014): 265.
19. Folwer H W, quoted in Garner B, The Elements of Legal Style (Oxford: Oxford University Press,
1991), 205–206.
20. Plain Language Action and Information Network, Federal Plain Language Guidelines, 45 (see
chap. 4, n. 4).
21. Dowell D, Kunins H, Farley T, “Letters: In Reply,” JAMA 310, no. 16 (2013): 1738.
23. Quill, “The Big Chill: Inserting,” 1 (see chap. 3, n. 16).
24. Feeney, “Antiviral Treatment of Hepatitis,” under “Introduction” (see chap. 3, n. 5).
25. Lai, “Long-Term Use of Zolpidem,” 593 (see chap. 4, n. 19).
26. Pelkowski, “Celiac Disease: Diagnosis and,” 99 (see chap. 1, n. 21).
27. Baron R, Kenny D, “The Moderator-Mediator Variable Distinction in Social Psychological
Research: Conceptual, Strategic, and Statistical Considerations,” Journal of Personality and
Social Psychology 51, no. 6 (1986), 1173.
28. Stedman’s Medical Dictionary, s.v. “Mediate.”
29. Zhang, “Association between Vaccination for,” 43 (see chap. 6, n. 6).
30. Ortega V, et al. “Effect of Rare Variants in ADRB2 on Risk of Severe Exacerbations and
Symptom Control During Long Acting β Agonist Treatment in a Multiethnic Asthma
Population: A Genetic Study,” Lancet Resp Med 2, no. 3 (2014), under “Background,” http://
www.thelancet.com/journals/lanres/articles/PIIS2213-2600(13)70289-3/fulltext.
31. Monti M, et al. “Willful Modulation of Brain Activity in Disorders of Consciousness,” N Eng J
Med 362 (2010), under “Results,” http://www.nejm.org/doi/full/10.1056/NEJMoa0905370.
32. Garattini S, Bertele V, “Europe’s Opportunity to Open Up Drug Regulation,” BMJ 340 (2010),
under “Transparency as a Means to Avoid Bias,” http://www.bmj.com/content/340/bmj.
c1578.
33. Carter, “B Cells in Health,” under “The Immune System and B Cells Form and Function” (see
Concept 1, n. 8).
34. Schaefer, “Urticaria: Evaluation and Treatment,” under “Etiology” (see chap. 6, n. 11).
84
85
CHAPTER 7
Statistical analysis of wseg scores
Everything should be made as simple as possible, but not simpler.
—Albert Einstein
Introduction
At the start of this book, we identified symptoms of medicus incomprehensibilis.
The tips in Chapters 1–6 addressed those symptoms related to low reading ease
and needless abstraction. The exercises in these chapters used medical journal
excerpts that showed at least one symptom. In the Exercise Key, we gave our revi-
sions and before-and-after wseg scores.
In this chapter, we give our analysis of these wseg scores. The original excerpts
had an average sentence length of 30.1 words, a 13.4 reading ease score, and an
18.6 grade level. Our revisions had an average sentence length of 14.1 words, a
57.9 reading ease score, and an 8.6 grade level. This analysis provides evidence you
can treat medicus incomprehensibilis effectively by applying the tips in this book.
Figures 7-1 through 7-4 show the breakdown of wseg scores. The top part of
each figure shows the raw scores.1 The bottom part shows the mean of all scores.
For Figures 7-2 through 7-4, we also show the ranges around the mean where
most scores fall. We show the range of plus or minus one standard deviation (µ ±
1σ), where 67% of scores fall. We also show the range of plus or minus two stan-
dard deviations (µ ± 2σ), where 95% of scores fall.
Analysis
T O TA L W OR D S ( w )
Figure 7-1 shows the distribution of total words. Total words for the originals
is shown in black. Total words for our revisions is shown in stripe. Total words
changed only slightly in our revisions.
The original excerpts had a mean of 34.2 words; our revisions had a mean of
34.8 words. But since we replaced long words with short ones where we could, our
revisions tend to take up slightly less space on the page.
85
86
35
30
25
Number of excerpts
20
15
10
0
0–10 11–20 21–30 31–40 41–50 51–60 61–70 71–80 81–90 91+
Number of words (W)
Original Revised
Figure 7-1 Distribution of total words (W)
S E N T E N C E L E N GT H (S)
Figure 7-2 shows the distribution of sentence lengths for the originals and our
revisions. The mean sentence length for the originals was 30.1 words. Sentence
lengths in the originals were widely distributed. The range of plus or minus one
standard deviation ran from 15.7 to 44.4 words per sentence. The range of plus or
minus two standard deviations ran from 1.3 to 58.8 words per sentence.
The mean sentence length for our revisions was 14.1 words. Sentence lengths
for our revisions were more narrowly distributed. The range of plus or minus one
standard deviation ran from 10.8 to 17.4 words per sentence. The range of plus or
minus two standard deviations ran from 7.6 to 20.7 words per sentence.
Overall, our revisions greatly reduced average sentence length. Even the high
end of our 95% range (20.7 words per sentence) was lower than the mean of the
originals (30.1 words per sentence).
R E A DI N G E A S E (E)
Figure 7-3 shows the distribution of reading ease scores for the originals and our
revisions. This distribution includes negative reading ease scores.2
87
70
60
50
Number of excerpts
40
30
20
10
0
0–10 11–15 16–20 21–25 26–30 31–35 36–40 41–45 46+
Average sentence length (S)
Original Revised
Figure 7-2 Distribution of average sentence length (S)
50
40
Number of excerpts
30
20
10
0
s
+
10
20
30
40
50
60
70
80
10
les
81
to
to
to
to
to
to
to
to
to
to
or
–9
11
21
31
41
51
61
71
9
0
–1
–2
Flesch Reading Ease score (E)
Original Revised
Figure 7-3 Distribution of reading ease scores (E)

88
The mean reading ease score for the originals was 13.4. Reading ease scores for
the originals were widely distributed. The range of plus or minus one standard
deviation ran from a reading ease score of −9.7 to 36.6. The range of plus or minus
two standard deviations ran from a reading ease score of −32.9 to 59.7.
The mean reading ease score for our revisions was 57.9. The reading ease scores
for our revisions were more narrowly distributed. The range of plus or minus one
standard deviation ran from a reading ease score of 45.5 to 70.3. The range of plus
or minus two standard deviations ran from a reading ease score of 33.1 to 82.7.
The low end of our 95% range (a 33.1 reading ease score) was much higher than
the mean of the originals (13.4).
G R A DE L E V E L ( G )
Figure 7-4 shows the distribution of grade levels for the originals and our revi-
sions. The mean grade level for the originals was 18.6. A non-native English
speaker may have an 18th-grade level knowledge of science, but their knowledge
of English may not be at the same level.
The grade levels for the originals were widely distributed. The range of plus or
minus one standard deviation ran from a grade level of 12.9 to 24.3. The range of
plus or minus two standard deviations ran from a grade level of 7.2 to 29.9.
The mean grade level for our revisions was 8.6. This grade level seems more
appropriate for a typical non-native speaker (e.g., think of a doctor from Europe,
90
80
70
Number of excerpts
60
50
40
30
20
10
0
0–3 4–6 7–9 10–12 13–15 16–18 19–21 22–24 25–27 28+
Flesch-Kincaid grade level (G)
Original Revised
Figure 7-4 Distribution of grade level scores (G)

89
Statistical Analysis of wseg Scores 89
Asia or Africa). Lowering the grade level by 10 school grades (8.6 vs. 18.6) should
cut reading time and improve reading comprehension for all readers.
The grade levels for our revisions were more narrowly distributed. The range
of plus or minus one standard deviation ran from a grade level of 6.7 to 10.6. The
range of plus or minus two standard deviations ran from a grade level of 4.7 to
12.6. The high end of our 95% range (a grade level of 12.6) fell far below the mean
of the originals (18.6).
Limitations
Of course, this analysis is not a formal study. Further research is needed (e.g., to
quantify the savings in reading time). The excerpts do not represent a random
sample of medical writing, nor do they represent medical writing as a whole.
Conclusion
We applied the tips in this book to the excerpts and greatly improved reading ease
and grade level. Our revisions reflect what can reasonably be achieved when writ-
ing in plain English. Using the tips in this book can help medical authors eradicate
Exercise 7. Putting the tips on reading ease and vivid language into practice

Review the symptoms of medicus incomprehensibilis listed in Table I-1 in the
Introduction. Read the excerpt from “The Frequency and Cost of Treatment
Perceived to be Futile in Critical Care,” out loud (Appendix 2, No. 2).
1. What symptoms of medicus incomprehensibilis do you see? Name at least five.

Describe the symptoms or give examples.
2. Revise using the tips on reading ease and vivid language.
Compare your revision to the Exercise Key in Appendix 3.
Notes
1. Scores have been rounded to the nearest integer.
2. We computed negative reading ease scores using the Flesch Reading Ease and Flesch-Kincaid
Grade Level formulas; see Wikipedia, s.v. “Flesch-Kincaid readability tests” https://en.wikipedia.
org/wiki/Flesch%E2%80%93Kincaid_readability_tests (accessed March 28, 2016). The wseg
score gave us the data we needed to compute the reading ease score, except for total number of
syllables. We found the total number of syllables, by using the grade level formula and solving
for the number of syllables.
90
91
concept
3
PRESENT LOGICAL
REASONING CLEARLY
A problem well-put is half-solved.—John Dewey
The last part of this book deals with clear logical reasoning. Much of the logic of
any medical research is set at the time the research project is approved. The out-
line for an article is set by the journal. Peer review checks the reasoning. The tips
we give here relate to presenting information clearly to help minimize medicus
incomprehensibilis.
The logic of a medical article is usually clear. Now and then, when it’s not, you
can still usually figure it out. But this may be your knowledge and experience as a
reader compensating for an unclear article.
Example—The scrambled owner’s manual

Imagine you’ve just presented your latest research at a medical conference in a
tropical location. As you drive your rental car around the island, a tire goes flat.
You decide to change it yourself, but you don’t know where to find the spare or the
jack. So you take out the owner’s manual.
Suppose the owner’s manual has good reading ease and uses vivid language,
but the directions are in the wrong order. (For example, it tells you how to tighten
a lug nut, before it tells you where to find the jack.) You puzzle over the directions.
Under these circumstances, if you’ve changed a tire before, you can probably
92
figure things out. But this is an instance of your knowledge and experience filling
in for the owner’s manual’s poor organization.
This example shows two things. First, how a narrative is organized is a separate
concept from reading ease or vivid language. Second, just because you can change
a flat tire using a scrambled owner’s manual, doesn’t mean the manual is well-
written; it ought to be written so anybody can change a flat tire. In the same way,
just because an expert can get the information they need from an article doesn’t
mean it is well-written. It should be written for the widest reasonable audience.
Why save logic for last?

Why do we address logical reasoning last after reading ease and vivid language?
For one thing, unclear logical reasoning makes the smallest contribution to medi-
cus incomprehensibilis. The logic of a medical journal article is usually clear. The
issues of logical flow we see now and then tend to be small. Perhaps a step of
reasoning is missing or ideas are presented in a confusing order.
The second reason is, when you improve reading ease and make your language
more vivid, it often helps reveal a problem with logical reasoning that may have
been hidden.
The third reason is, it’s more complicated. Logical reasoning is harder to
discuss, since you need to look at bigger chunks of text—several sentences or
paragraphs—not just one sentence. Because logic deals with both substance and
form, you need to understand the science. You also need to understand what the
author is trying to say.
Improving the flow of logical reasoning takes time and thought, but it’s worth
it. When you improve logic, you improve both style and content.
In the next three chapters, we give tips on presenting logical reasoning clearly.
8. Organize your narrative in a way that’s helpful for your reader

9. Choose a clear narrative pathway
10. Forge a strong chain of logical reasoning
Overview of Concept 3 exercises

Since analyzing logic involves looking at bigger chunks of text, Concept 3 uses
fewer exercises, and each exercise takes more thought. In Chapter 8, we look at
paragraphs and ask you to consider how you might present the information in a
way that is more helpful to the reader.
In Chapters 9 and 10, we look at an excerpt from Mathematical Modeling of
Kidney Transport. We ask you to consider how you might adapt it reach a wider
audience.
93
CHAPTER 8
Organize your narrative in a way
that’s helpful for your reader
Good paragraphs have unity. They have topic sentences that announce
the idea to be developed in the paragraph, and then they stick to that
idea. And headings can powerfully reinforce the unity of your para-
graphs.—Bryan Garner1
A good narrative organizes information in a way that helps the reader. When a
narrative fails to organize information well, it adds to medicus incomprehensibilis.
One failure we occasionally see is the overly long paragraph. Another is that
a narrative presents two-dimensional data as one-dimensional standard prose.
Either one of these writing habits places an unnecessary burden on the reader to
analyze and interpret the information. This chapter covers simple strategies for
treating these problems.
A. Introduce and develop one idea in each paragraph

A paragraph should start with a sentence that suggests the topic or helps with
transition from the previous paragraph. Or else, it may tell about the new para-
graph’s function as part of the whole. If you start a paragraph with a good heading
or topic sentence, it can help your reader quickly grasp that idea.2
Dividing a narrative into good paragraphs helps a reader see the progression of
ideas. If a single paragraph goes on too long, it may shift the burden of seeing the
progression of ideas over to the reader. The reader can no longer simply read the
paragraph; instead, they have to study it.
A good paragraph heading or topic sentence is the reader’s friend. If they ever
want to look back to find something, they can do so quickly by looking at the
headings or topic sentences.
How long should a paragraph be? Some authors advise that a writer should
strive for an average paragraph of no more than 150 words, with a mixture of
short and long paragraphs.3
93
94
Exercise 8.A. Introduce and develop one idea in each paragraph

For each excerpt:
1. Read it out loud. Do you understand it just by reading it, or do you need to
study it?
2. Does it cover one idea or multiple ideas?
3. Without re-writing the text, split up the long paragraph into shorter para-
graphs no more than 150 words each, covering just one idea or topic. Write a
heading or topic sentence for each new paragraph.
4. In your view, do shorter paragraphs with headings or topic sentences make
this excerpt easier to read? Why or why not?
Compare your answers with the Exercise Key in Appendix 3.
1. In a majority of cases, metabolism that is mediated by cytochrome

ples include clopidogrel (metabolized by CYP3A4), cyclophosphamide
(metabolized by CYP2B6) and tamoxifen (metabolized by CYP2D6).
The major pathway of codeine consists of glucuronidation and N-
demethylation, whereas the CYP2D6-mediated O-demethylation to
produce morphine is a minor reaction. Nevertheless, the latter is a cru-
cial step in bioactivation, since the affinity of codeine for the µ-opioid
receptor is only 1/200 to 1/3000 that of morphine. Previous studies
have shown that the effects of codeine—analgesic, respiratory, psy-
chomotor, and miotic—are markedly attenuated in people with poor
metabolism of CYP2D6. On the other hand, people with ultrarapid
metabolism, such as the patient described by Gasche et al. in this issue
of the Journal, produce greater amounts of morphine from codeine
and therefore may experience exaggerated pharmacologic effects in
response to regular doses of codeine. Similar effects, albeit less dra-
matic, have been described in patients with ultrarapid metabolism of
CYP2D6 in response to routine doses of hydrocodone or oxycodone,
which are other opioids requiring CYP2D6 mediated activation. These
reports clearly illustrate the effect of CYP2D6 genetic polymorphisms
on the action of codeine, ranging from virtually no effect in patients
with poor metabolism to severe toxic effects in those with ultrar-
apid metabolism. To put these observations into perspective, these
extremes of response might be relevant for some 10 to 20 percent of
whites who have phenotypes associated with either poor metabolism
or ultrarapid metabolism.4 (wseg = 269/29.8/10.9/18.4)
2. The [pulmonary artery catheter] PAC, today still considered the
clinical gold standard for CO estimation, gives three fundamental bits
95
Organize Your Narrative in a Way That's Helpful 95
of hemodynamic information: CO, pulmonary and cardiac filling pres-

sures and mixed SvO2. The PAC is considered the clinical gold stan-
dard or reference method for CO estimation and every new tool built
for CO estimation has to be compared with PAC in validation studies.
The technique is based on the injection of an ice-cold solution into the
right atrium (proximal port—prox injectate—blue lumen). The change
in blood temperature is measured in the pulmonary artery by a therm-
istor placed proximally to the tip of the catheter. The thermodilution
curve is used for CO estimation by means of the Steward-Hamilton
formulation. The measurement is repeated at least three to five times
(in order to compensate for variations induced by the respiratory
cycle) and the average value is then calculated. An “inverted” ther-
modilution curve can be obtained by worming [sic] the blood with a
thermal filament (VigilanceTM, Edwards Lifesciences, Irvine, CA, USA)
or a thermal coil (OptiQTM, ICU Medical, San Clemente, CA, USA). This
system allows for a semi-continuous CO measurement by display-
ing average CO values for the previous 10 min and limiting, for this
reason, the effects of arrhythmias and other compounding factors.
Although the CO value is frequently up-dated, this type of monitor-
ing is not continuous (beat-by-beat) and is therefore less accurate and
rapid in detecting hemodynamic instability. Major limiting factors for
CO estimation with any type of thermodilution is the occurrence of
tricuspid regurgitation and intracardiac shunts. In fact, prolonged
indicator transit times or indicator recycling may lead to errors in CO
estimation. SvO2 suggests whether or not cardiac output is adequate
in a patient since it provides a useful indication about the adequacy
of tissue oxygenation in specific conditions of metabolic activity.
When SvO2 decreases below normal values (70–75%) in the presence
of normal arterial oxygen saturation and without anemia, it means
that CO is inadequate and measures aimed at increasing DO2 should
be promptly implemented. SvO2 can be measured either continuously
(fiberoptic fibers) or intermittently by withdrawing a mixed venous
blood sample from the distal lumen of the PAC. The PAC measures: the
pulmonary artery pressure which represents right ventricular after-
load; right atrial pressure (RAP/CVP, PAOP) have been demonstrated
to be less reliable than dynamic indicators of fluid responsiveness
(pulse pressure variations PPV, systolic pressure variation SPV and
stroke volume variation SVV). Nonetheless, when static parameters
are particularly low, they can be considered to be as reliable as dynamic
ones. Over recent years, the use of PAC has decreased significantly for
two main reasons: firstly, several randomized and non-randomized
studies have not demonstrated an improvement in patient outcomes
when therapies were guided by PAC and secondly nowadays the PAC
is the most invasive tool for hemodynamic monitoring currently used
96
in ICUs or ORs. In 1996, Connors and co-workers published in JAMA

a prospective cohort study in which the authors used case-matching
multivariable regression modeling techniques and a propensity score.
The results showed that pulmonary artery catheterization in critically
ill patients was associated with an increased risk of death (odds ratio
1.24; 95% confidence interval 1.03–1.49) as well as a prolonged length
of stay and increased resource utilization. After this shocking paper,
a litany of randomized trials aimed at confirming these results was
performed in ICUs. The net results of this great amount of data cor-
responded to an increased suspicion that PAC-guided treatment may
not be superior to non-PAC-guided treatment and a marked decrease
in PAC use worldwide has become clear during the last 10–15 years.
In the USA, Wiener et al. reported a 65% decrease in PAC use between
1993 and 2004 and in Canada, Koo et al. reported a more than 50%
decrease between 2002 and 2006. A great number of papers has been
published by opinion leaders and general considerations about the
use of PAC and its benefits. The doubts regarding its usefulness can be
listed as follows.5 (wseg = 667/27.7/18.5/17.0)
B. Present two-dimensional data in a table,

chart or graph
In any language, prose is one-dimensional. We write a string of words arranged in
a line according to the order in which we speak them. English writes from left to
right. Other languages write from right to left or top to bottom.
Unlike standard prose, some data is inherently two-dimensional. We can often
understand a collection of data better if it’s presented in a table, chart or graph.
This often reduces the number of words, improves reading ease, and makes the
data easier to grasp. (For example, when data is presented in a table, we some-
times choose to scan data down the columns, rather than across the rows.)
Let’s consider an example:
Vasectomy reversal techniques involve reanastomosis of the testicular

and prostatic vasal ends (vasovasostomy) or connecting the vas to the
epididymis (vasoepididymostomy). Vasovasostomy patency rates have
been reported between 75% and 86%; pregnancy rates range from 45%
to 70%. Vasoepididymostomy, a technique used in the presence of epi-
didymal obstruction, has patency rates between 31% and 92%, and
pregnancy rates between 10% and 50%. Repeat attempts at microsur-
gical vasectomy reversal appear less successful than first attempts with
patency rates between 75% and 79%, and pregnancy rates between
32% and 43%.6 (wseg= 88/22.0/7.2/16.6)
97
This subject, vasectomy reversal and pregnancy, has a high level of human inter-
est, but the narrative sounds dull and reading ease is low. Why? It’s partly due
to long sentences and essential medical terms. It’s also partly due to the fact
that the data is presented in one-dimensional prose, which makes the reader
have to work harder to understand. The data becomes easier to grasp if we orga-
nize it into a table. For example, we might summarize the same data as follows:
Vasectomy reversal techniques involve re-attaching the testicle and

prostate ends of the vas (vasovasostomy) or connecting the vas to
the epididymis (vasoepididymostomy). Table 8-1 gives data on the
reported outcomes for operations that use the different reversal
techniques.
Table 8-1. Outcomes for vasectomy reversal
Success rate (%)
Procedure name Procedure/indication Vas freely open Pregnancy

Vasovasotomy Reconnect vas (prostate 75–86 45–70
and testicle ends)
Vasoepididymostomy Connect vas to epididymis, 31–92 10–50
used when epididymis is
blocked
Repeat attempts Previous attempt to 75–79 32–43
reconnect vas failed
The table helps the reader see three different types of information at a glance:
(1) the procedures, (2) the possible success outcomes, and (3) the data that quan-
tify the success rate.
Of course, creating a table requires the author to do more work. A table takes
up more space on the page. But if you consider the value of the time saved for
hundreds or thousands of readers, all busy professionals, the extra time and paper
are certainly worth it.
For a good short discussion of creating effective tables, charts and graphs, we rec-
ommend, The AMA Manual of Style,7 or The Craft of Research by Booth, Colomb and
Williams.8 For a more extensive discussion, consider the works of Edward Tufte.
S UM M A RY
Presenting two-dimensional data in one-dimensional form often adds to medicus

incomprehensibilis. Presenting data in two-dimensional form, as a table, chart or
graph, can help treat the problem.
98
Exercise 8.B. Present two-dimensional data in a table, chart or graph

For each sentence below:
1. Read it out loud. Underline any data you think might be clearer if presented in
two-dimensional form.
2. Create a table or chart to present the data in two-dimensional form. Revise the
remaining text to improve reading ease.
1. All 525 study participants, who were randomized to receive vareni-
cline or placebo, had been diagnosed with major depressive disorder
and were being treated with antidepressant drugs at a stable dose or
had been successfully treated for depression within the past 2 years.
At 9 to 12 weeks, 35.9% of those who received varenicline quit vs.
15.6% taking placebo; at 40 weeks, 20.3% of the varenicline group
had quit compared with 10.4% of the placebo group. Depression and
anxiety did not increase in either group, but the researchers cau-
tioned that their findings may not apply to smokers whose depres-
sion isn’t successfully treated.9
2. What about health in Scotland? According to the UK’s national
statistical office, healthy life expectancy was 59·8 years for men and
64·1 years for women in Scotland during 2008–10, 4·6 years and
2·3 years fewer than for men and women in England, respectively.
According to the British Heart Foundation, 35% of Scottish men and
30% of women have high blood pressure; alcohol use is one noticeable
contributor to ill health in Scotland, with up to 50% of men and 30%
of women exceeding guidelines for drinking.10
Conclusion
A good narrative organizes information in a way that helps the reader. In this
chapter, we talked about two simple ways to organize information. Make sure a
paragraph is not overly long and deals with just one subject. Use a heading or topic
sentence to help preview the content for the reader. Organize two-dimensional
data using a table, chart or graph. This allows the reader to grasp at a glance how
each idea relates to the others.
In the next chapter, we look at narrative pathway.
99
Notes
1. Garner B, Securities Disclosure in Plain English (Chicago: CCH Incorporated, 1999), 64.
2. Strunk, The Elements of Style, 15–17 (see chap. 4, n. 2).
3. Garner, Securities Disclosure in Plain, 61–73.
5. Ramagnoli S, “Circulatory Failure: Exploring Macro-and Micro- Circulation,” Trends in
Anaesthesia and Critical Care 3 (2013), under “The Pulmonary Artery Catheter [PAC],” http://
www.trendsanaesthesiacriticalcare.com/article/S2210-8440(13)00020-8/fulltext.
6. Rayala, “Common Questions about Vasectomy,” 760 (see chap. 2, n. 7).
7. Iverson C, et al. ed. AMA Manual of Style, 10th ed. Oxford: Oxford University Press, 2007.
8. Booth W, Colomb G, Williams J, The Craft of Research, 3rd ed. (Chicago: University of Chicago
Press, 2008); see chap. 15 on “Communicating Evidence Visually.”
9. Slomski A, “Clinical Trials Update: Depression Remains Stable in Smokers Taking Varenicline
to Quit,” JAMA 310, no. 16 (2013): 1165.
10. “Scotland: Towards a Healthy and Interdependent Future,” Lancet 384, no. 9947 (2014),
http://www.thelancet.com/journals/;ancet/article/PIIS0140-6736(14)61614-7/fulltext.
100
101
CHAPTER 9
Choose a clear narrative pathway
If he would inform, he must advance regularly from Things known to
things unknown, distinctly without Confusion, and the lower he begins
the better. It is a common Fault in Writers, to allow their Reader too
much knowledge: they begin with that which should be in the Middle,
and skipping backwards and forwards, ‘tis impossible for anyone but he
who is perfect in the Subject before, to understand their Work, and such
an one has no Occasion to read it.—Benjamin Franklin1
Why does an expert in their field talk over everybody’s head? Why can’t they talk
about their subject in a simple way? Is it just because their ideas are too advanced
or complex? Perhaps. But sometimes when an author understands their subject
well, they become blind to what other people do or don’t understand. They over-
estimate the reader’s knowledge. They misjudge the widest reasonable audience,
defining it too narrowly, composed of people just like themselves. As a result, they
may do several things that confuse the reader.
1. They fail to start by talking about “things known.”

2. They fail to start by anchoring their discussion in the real world.
3. They fail to follow a clear narrative pathway.
4. They make an abrupt transition between concrete and abstract or vice versa.
Narrative pathway is the direction of a narrative, or a conceptual program for orga-

nizing a narrative. Centuries ago, Benjamin Franklin saw that some people who
wrote about science had a problem organizing their ideas. We see the same thing
today. In this chapter, we give tips to avoid common problems and help you create
a clear narrative pathway.
A. Start with things known

A good narrative often starts by stating things commonly known. Before talk-
ing about new discoveries developed through research, a medical article starts by
101
102
talking about prior research and current knowledge and practice. For example,
the article, “Effect of Aspirin and Antiplatelet Drugs on the Outcome of the Fecal
Immunochemical Test,” begins by reviewing facts about colorectal cancer and
common screening techniques.
Colorectal cancer (CRC) is the third most common cancer worldwide

and the second leading cause of cancer-related deaths. Evidence from
several studies has indicated that CRC screening is effective and cost-
effective in average-risk populations. Recommended CRC screening
strategies fall in 2 broad categories: stool tests that primarily detect
cancer, which include detection of occult blood or exfoliated DNA, and
structural tests, such as flexible signoidoscopy, colonoscopy, and com-
puted tomographic colonography, which are effective in detecting both
cancer and premalignant lesions.2
Exercise 9.A. Start with things known

Review the excerpt Mathematical Modeling of Kidney Transport in Appendix 2.
Do you think it starts by talking about concepts familiar to the widest

reasonable audience? Why or why not?
Compare your answer to the Exercise Key in Appendix 3.
B. Start by anchoring your discussion in

the real world
Starting with things known often involves starting in the real word. In Chapter 5,
we saw how a sentence that uses a concrete subject is more vivid and easier to
understand. The same concept applies to logical reasoning. People usually find the
real world easier to grasp than an abstract idea. Science writing often proceeds
from talking about research data (real-world facts) to analyzing them and drawing
conclusions (abstract analysis).
Medical articles start with an abstract that summarizes the research. A reader
often reads the abstract to decide whether to read the article. Therefore, it helps if
the abstract is stated in real-world terms.
Consider two excerpts from the abstract of a recommendation statement by
the US Preventative Services Task Force (Table 9–1). The statement covers the
issue of whether a doctor should screen vision for an older patient. The original
statement is not hard to understand, but it sounds formal and abstract. In par-
ticular, the term “visual acuity” (literally, sharpness or clearness of seeing) seems
an abstract way to describe how well a patient sees things in the real world.
103
Choose a Clear Narrative Pathway 103
Table 9-1. Start by anchoring the discussion in the real world
Originali Revised
DESCRIPTION DESCRIPTION OF THE ISSUE
Update of the US Preventive Services Should a doctor screen the vision of a
Task Force (USPSTF) recommendation patient over 65? (wseg = 11/11.0/72.6/5.8)
on screening for impaired visual acuity in
older adults. (wseg = 19/19.0/36.1/12.9)
RECOMMENDATION RECOMMENDATION (UPDATE)
The USPSTF concludes that the current None. We don’t think there is enough
evidence is insufficient to assess the evidence to decide this. (wseg = 11/5.5/
balance of benefits and harms of 85.8/2.6)
screening for impaired visual acuity in
older adults. (wseg = 26/26.0/37.2/14.5)
US Preventive Services Task Force, “Screening for Impaired Visual Acuity in Older Adults: US Preventive
i
Services Task Force Recommendation Statement,” JAMA 315, no. 9 (2016): 908.
We can make this statement clearer by revising it to state the issue in shorter,
real-world terms. Getting the first step right, by stating the issue in real-world
terms, helps make everything that follows easier to understand.
Though it’s usually best to start a narrative by talking about the real world,
there are a few exceptions. An article could start with a well-known abstract
concept and move to more concrete discussion. For example, an article could
start by taking about the theory of evolution, a well-known abstract concept,
and go on to talk about how it applies in immune therapy used to treat a patient
with HIV.
Exercise 9.B. Start by anchoring your discussion in the real world

Look again at the excerpt Mathematical Modeling of Kidney Transport in Appendix 2.
Does the narrative start by talking about the real world? Or does it seem
abstract?
Compare your answer to the Exercise Key in Appendix 3.
C. Choose a good narrative pathway

If a good narrative starts with things known and in the real world, where should
it go from there? The key is to pick a good narrative pathway, make it clear to the
104
Table 9-2. Examples of common narrative pathways

Size Proceeds from big to small or vice versa.
List A set of items that have something in common.
Time Starts with earlier events and proceeds to later events.
Process Actions in a particular order: step 1, step 2, etc.
Starts by describing general information and then gives
Pyramid
more detailed information.
Problem/solution Starts by describing a problem and then gives a solution.
Starts by describing the status quo, tells what’s wrong
Status quo/alternative/
with the status quo, presents an alternative, and
difference
evaluates the difference.
Starts by describing real-world things or actions (e.g.,
Real world/abstract
data) and then moves to more abstract analysis.
reader, and stick to it. It often helps a reader follow a narrative if they know where
it is going. Table 9-2 shows examples of common narrative pathways.
How can you make your narrative pathway clear to the reader? It often takes
only a simple cue. Consider the paragraph heading: “How Common is Barrett
Esophagus and What Are the Risk Factors?”3 From this statement, the reader
learns what types of information will follow. But a reader might be confused if
this same paragraph went on to talk about the history of Barrett’s Esophagus or
ways to treat it.
Exercise 9.C. Choose a good narrative pathway

Look again at the first two paragraphs of the excerpt from Mathematical Modeling
of Kidney Transport in Appendix 2.
1. What narrative pathways does the narrative follow?

2. Does it state them explicitly? Or are they clear from context?
3. Does the excerpt follow each narrative pathway consistently?
D. Make a smooth transition between concrete

and abstract
One big challenge of medical writing is to make a smooth transition between
concrete and abstract, or vice versa. It is often a smooth and natural transition
to move from the real world to abstract analysis. Most research articles have
105
Choose a Clear Narrative Pathway 105
separate sections for study design, patients or study subjects, study outcomes, and
interpretation of results. Within each section, it is often clear from context whether
the discussion deals with the real world or abstract ideas. Keeping real-world
and abstract ideas in separate sentences or paragraphs can also help make the
transition clear.
Exercise 9.D. Make a smooth transition between concrete and abstract

Look again at the first two paragraphs of Mathematical Modeling of Kidney
Transport in Appendix 2.
1. What real-world things and actions does the narrative mention? What abstract
ideas does it mention? Does the narrative make a smooth transition between
concrete and abstract?
2. In what way are a rat kidney and a human kidney enough alike that it makes
sense to compare them? Does the narrative explain this?
3. In what way are the kidneys of a rat, human, elephant and whale enough alike
we can generalize about an abstract “mammal kidney?” Does the narrative
explain this?
4. Write a paragraph that tells why we can generalize about a “mammal kidney.”
Compare the number of nephrons in a rat kidney and a human kidney.
Conclusion
In this chapter, we talked about choosing a clear narrative pathway. This usually
involves starting with things known and anchoring the discussion in the real
world. From there, choose a good narrative pathway, make it clear to the reader,
and stick to it. Make a smooth transition between concrete and abstract.
We develop these ideas further in the next chapter, where we give tips on forg-
ing a strong chain of logical reasoning.
Notes
1. Quoted in Williams, Style: Lessons in Clarity, 74 (see Preface, n. 8).
2. Bujamda, “Effect of Aspirin and,” 683 (see chap. 1, n. 27).
3. Zimmerman, “Common Questions about Barrett’s,” 92 (see chap. 2, n. 16).
106
â•‡ 107
CHAPTER 10
Forge a strong chain of logical reasoning
Looking back, I think it was more difficult to see what the problems were
than to solve them.—â•‰Charles Darwin1
Why doesn’t an expert write clearly? In this chapter, we give the second part of our
answer. An expert, someone “too close” to their subject, doesn’t always present a
strong chain of logical reasoning. Sometimes they leave out a step of reasoning
that seems obvious to them or present ideas in a confusing order (Table 10-â•‰1).
Many subjects would interest a wider audience, if only the author would explain
the subject step by step in the right order. In this chapter, we give tips on forging a
strong chain of logical reasoning.
A.â•‡ Explain each step of reasoning

A good narrative explains each step of reasoning. Somebody too close to their sub-
ject may skip a step of reasoning they think seems obvious. But what seems obvi-
ous to an expert may not be obvious to the widest reasonable audience. Nobody
ever tries to skip a step of reasoning, but it happens. Peer review may not always
catch a missing step, especially when the reviewer is an insider, and the article’s
reasoning is clouded by medicus incomprehensibilis.
How do you spot a missing step of reasoning? The best way is to write in plain
English, so any gap in reasoning becomes easier to see and fix.
Exercise 10.A.â•‡ Explain each step of reasoning

Look again at Mathematical Modeling of Kidney Transport in Appendix 2.
107
108
Table 10-1. An expert often leaves out steps of reasoning
They talk about… but fail to first explain…

how to solve a problem what the problem is or why it is important
the problem in abstract terms the problem in real-world terms
details or steps of reasoning things to be proven or conclusions
a formula or equation in symbols the formula or equation in words
The narrative starts by talking about the major parts of a kidney, the cortex
and medulla. It goes on to talk about a nephron and its parts.
1. Where is a nephron located in relation to the cortex and medulla?

2. Does the excerpt explain this? Or has it left out a step of reasoning?
B. State the problem before you solve it

A medical research article often involves answering three questions: What? Why?
and How?
W H AT ?
What is the problem? What is the question the research project is designed to
address? Most medical research sets out to answer a question (e.g., Does treat-
ment “A” work better than treatment “B?” Does the new medicine work better than the
old one?)
W HY ?
Why is the question or problem important? For a doctor or other medical scien-
tist, a problem becomes important when it affects the health or treatment of a
patient.
HO W ?
How did the research seek to answer the question or solve the problem? How was
the experiment or trial conducted?
It can be harder to follow a narrative that mixes up these three questions or
leaves one of them out. Some authors try to tell what they did and how they did it in
the same sentence. Some explain how they solved a problem without first stating
109
Forge a Strong Chain of Logical Reasoning 109
clearly, what the problem is, or why it is important. Some authors explain how
they carried out their research without first clearly stating what the question was
their research was designed to answer. A good statement of the research question
or problem goes a long way toward helping a reader understand the solution.
A medical research article starts with an abstract that addresses the what,
why and how questions. But an abstract doesn’t always frame these questions as
clearly as it might. For example, consider the statements of “Importance” and
“Objective” from a research article abstract in Table 10-2.
Do these statements answer the what and why questions as clearly as possible?
We can understand these statements with a little study. But we can revise to state
the research question, and tell why it is important, more simply and clearly. No
doubt the authors had to write their abstract to fit the journal’s format, which
required them to state their research “objective” rather than to frame the research
question.
The main text also addresses the research question, but clouds it with details of
research method and analysis.
The primary objective of this trial was to determine whether degludec/

liraglutide was noninferior to up-titration of glargine in change from
Table 10-2. State the problem before you solve it
Originali Revised
Importance Importance
Achieving glycemic control remains For a patient with type 2 diabetes, it can be
a challenge for patients with type 2 hard to keep blood sugar under control, even
diabetes even with insulin therapy. if they take insulin. (wseg = 22/22.0/65.2/9.6)
(wseg = 16/16.0/26.6/13.5)
Objective Objective—Issue addressed
To assess whether a fixed ratio of A patient has type 2 diabetes. Their doctor
insulin degludec/liraglutide was treats them using insulin glargine and
noninferior to continued titration metformin.
of insulin glargine in patients with If this doesn’t work to control their blood
uncontrolled type 2 diabetes treated sugar, the common next step is to raise their
with insulin glargine and metformin. insulin dose, as needed. This study poses the
(wseg = 31/31.0/3.4/20.4) question: Would it work just as well to use
a fixed ratio of insulin degludec/liraglutide?
(wseg = 56/14.0/67.2/7.3)
Lingvay I, et al. “Effect of Insulin Glargine Up-titration vs Insulin Degludec/Liraglutide on Glycated
i
Hemoglobin Levels in Patients With Uncontrolled Type 2 Diabetes—The DUAL V Randomized Clinical
Trial,” JAMA 315, no. 9 (2016): 898.
110
baseline HbA1c level in patients with uncontrolled type 2 diabetes

treated with glargine and metformin. If the primary objective was
met, secondary objectives were to assess whether degludec/liraglutide
was statistically superior compared with glargine in change from base-
line of HbA1c level, body weight, and rate of confirmed hypoglycemia.2
(wseg = 70/35.0/4.5/21.3)
Part of what makes this excerpt hard to understand is that it covers both what
(the research question) and how (research method and analysis). The research
question is: Does degludec/liraglutide work as well as insulin glargine with metformin
to help control a patient’s diabetes?
The research method involves how the researchers designed the clinical trial,
and analyzed the data, in order to answer the research question. This involves talk
of: up-titration, baseline HbA1c level, statistically superior, body weight, and rate of
confirmed hypoglycemia. The statement of the research question would be clearer if
the discussion of what and how were kept separate.
Exercise 10.B. State the problem before you solve it

Look again at Mathematical Modeling of Kidney Transport in Appendix 2.
1. What happens if a person’s kidney stops filtering blood as it should?

2. Why is it important to have a math model of how the kidney filters blood?
3. Does the excerpt help you to answer these questions?
C. Say it in words before you say it in symbols

Medical writing sometimes presents an equation or formula in symbols without
ever saying, in words, what the equation or formula represents. Consider the
equation, 2223 –(20 × 60) = 1023. Stated in symbols alone, it’s impossible to see
why this equation or its solution are important.
But if we state the problem in words, its importance becomes clear and vivid.
The HMS Titanic sailed on its maiden voyage in April 1912 with 2223
passengers and crew aboard. The ship had 20 lifeboats, and each life-
boat could carry 60 people. If the ship were to strike an iceberg and
sink quickly, far from any source of help, how many people would have
to be left behind?
111
Table 10-3. Words and symbols help make a smooth transition
Words Words & symbols Symbols

We need to prepare a field hospital to 7 tents × 8 cots =? patient 7 × 8 = 56
care for patients during a sudden Ebola capacity per truckload
outbreak in a remote location. One truck
can carry seven tents, with eight cots for
each tent. How many patients can we
accommodate using each truckload of
equipment?
M AK I NG A S MO O T H T R A N SI T ION
Stating a problem in words makes it vivid and easy to understand. Stating it in

symbols makes it easy to solve. Using a combination of words and symbols can
sometimes help make a smooth transition between the two. (Table 10-3)
A schematic diagram sometimes serves as an intermediate step between the
real world and abstract analysis. Thus, it can help make a smooth transition. For
example, it is common for a schematic diagram of the circulatory system to show
a simplified drawing of the heart, veins and arteries. Arrows show the direction
of blood flow. Areas where oxygenated blood flows are commonly colored in red.
Areas where de-oxygenated blood flows are commonly colored in blue. As such,
some aspects of the diagram depict the real-world, other aspects seem more like
abstract ideas.
One common way to make a smooth transition is to state a problem three
times: (1) once in real-world terms (in words), (2) again in terms of math or other
scientific concepts (in words), and (3) a third time in math or other scientific sym-
bols. Taking care to make a smooth transition can go a long way towards present-
ing a logical argument clearly.
For example, suppose a patient has hip surgery and the doctor wants to pre-
scribe oral pain medicine for them. How does a doctor figure the right dose?
A good description may involve stating the problem three times (Table 10-4).
In a journal, this narrative might look like this:
What is the right oral dose of medicine to relieve a patient’s pain? It

is the dose that achieves a concentration of medicine in the patient’s
plasma that is both safe and effective. This is called, the target thera-
peutic concentration (TTC). To be effective, the TTC needs to be higher
than, or equal to, the minimum effective concentration (MEC). To be
112
Table 10-4. Three ways to describe dosing for pain medicine
Oral dose Concentration in plasma Math

What is the right It is the dose that achieves a We can express this
oral dose of concentration of medicine in the idea in the equation:
medicine to relieve patient’s plasma that is both safe and MEC ≤ TTC ≤ MSC.
a patient’s pain? effective. This is called, the target
therapeutic concentration (TTC).
To be effective, the TTC needs to be
higher than, or equal to, the minimum
effective concentration (MEC). To
be safe, the TTC needs to be lower
than, or equal to, the maximum safe
concentration (MSC).
safe, the TTC needs to be lower than, or equal to, the maximum safe
concentration (MSC). We can express this idea in the equation: MEC ≤
TTC ≤ MSC.3 (wseg = 88/14.6/65.0/7.8)
Exercise 10.C. Say it in words before you say it in symbols

Look again at the excerpt from Mathematical Modeling of Kidney Transport in
Appendix 2. The excerpt presents three equations.
1. Does the narrative tell in words what each equation represents?

2. Does it otherwise explain them well enough for the widest reasonable audi-
ence to understand?
Conclusion
In this chapter, we gave tips to help you forge a strong chain of logical reason-
ing: Explain each step of reasoning. State the problem before you solve it. Say it in
words before you say it in symbols.
113
Notes
1. Quoted in Williams, Style: Lessons in Clarity, 185 (see Preface, n. 8).
2. Lingvay I, et al. “Effect of Insulin Glargine Up-titration vs Insulin Degludec/Liraglutide on
Glycated Hemoglobin Levels in Patients With Uncontrolled Type 2 Diabetes—The DUAL V
Randomized Clinical Trial,” JAMA 315, no. 9 (2016): 898–899.
3. Linares O, et al. “Personalized Oxycodone Dosing: Using Pharmacogenetic Testing and Clinical
Pharmacokinetics to Reduce Toxicity Risk and Increase Effectiveness,” Pain Med 15, no. 5 (2014).
114
115
Afterword—Can things ever change?

The formulation of a problem is often more essential than its solution,
which may be merely a matter of mathematical or experimental skill.
To raise new questions, new possibilities, to regard old questions from
a new angle, requires creative imagination and marks real advance in
science.—Albert Einstein
In the 21st century, English is the global language of medical science. Reaching
the widest reasonable audience requires writing in a way that is understandable.
In this book, we showed how medicus incomprehensibilis mostly stems from
needless grammatical complexity. We identified several over-used writing habits
that are symptoms of this complexity. We showed how you can use a small collec-
tion of plain English writing tips to treat medicus incomprehensibilis and improve
reading ease, vividness and logical flow.
Change comes when people share a vision. We hope this book has provided
a clear vision of what plain English is, why it’s important, and how to use it to
improve your medical writing.
115
116
117
Appendix 1
ENGLISH SPEAKERS AROUND THE WORLD
Introduction
Anybody who wants to write for the widest reasonable audience needs to consider
the world’s non-native English speakers. This appendix surveys English speak-
ers around the world, including both native and non-native speakers. We pres-
ent data on English speakers in general, since it gives a rough idea of where the
world’s English-speaking doctors live.
Non-native English speakers constitute the majority of the world’s total English
speakers. In predominantly English-speaking countries, they are an important
minority. Since doctors are among the best-educated people in any country, it
stands to reason they are over-represented among each country’s English speak-
ers. For example, if 10% of India’s citizens speak English, it stands to reason far
more than 10% of Indian doctors speak English. (We think the number is closer to
100%.) The same probably holds true in any other country where learning English
is considered part of a “good education.”
What does it mean to be an “English speaker?”

The language skills of English speakers vary. A non-native speaker who has lived
all their life in England may speak English better than most native speakers.
Other non-native speakers may speak English well, but not as well as a native
speaker. For example, a German doctor may have good scientific training and read
and speak English well, but they might not speak English as well as a high-school
graduate from Australia.
The world’s top 25 English-speaking countries

Table A1-1 shows a list of the world’s top 25 English speaking countries. More
than half the world’s population lives in these countries.
117
118
Table A1-1. World’s top 25 English-speaking countries (population in millions)i
Total Population English On which continent?

(millions) Speakers
Asia North Europe Africa Australia South America
(millions) % America
1 United States 317 298 94% 298
2 India 1,210 125 10% 125
3 Pakistan 188 92 49% 92
4 Nigeria 156 83 53% 83
5 UK 64 64 100% 64
6 Philippines 100 64 64% 64
7 Germany 81 52 64% 52
8 Bangladesh 163 30 18% 30
9 Canada 33 28 85% 28
10 Egypt 83 28 34% 28
11 France 65 26 40% 26
12 Italy 60 20 33% 20
13 Australia 21 17 81% 17
14 Thailand 63 17 27% 17
15 South Africa 53 16 30% 16
16 Netherlands 17 15 89% 15
17 Poland 39 14 37% 14
119
18 Nepal 30 14 47% 14
19 Turkey 71 12 17% 12
20 Iraq 32 11 34% 11
21 Brazil 205 11 5% 11
22 Spain 47 10 21% 10
23 China 1,200 10 1% 10
24 Sweden 10 8 81% 8
25 Kenya 43 8 19% 8
Total 4,350 1,073 25% 375 326 209 135 17 11
i
We took data for Population and English Speakers from Wikipedia, s.v. “List of Countries by English-Speaking Population,” https://en.wikipedia.org/wiki/List_of_
countries_by_English-speaking_population (accessed May 5, 2016) and rounded to the nearest million. We figured % by dividing English Speakers by Population. We
included Turkey in the Asia column, though partly in Europe.
120
English is strongly represented on every continent except South America and

Antarctica. The list in Table A1-1 includes nine countries in Asia, eight in Europe,
and four in Africa. Asia now has more English speakers than any other continent,
and the number of English speakers there seems poised to grow in the future.
U S A , UK , C A NA DA A N D AU S T R A L I A
These large countries, where most people speak English, are home to 407 million
people or 5.5% of the world’s population.1 Within these countries, about 14% of
the total population, or 59 million people, are non-native English speakers. Taken
together, these countries have 247 medical schools (Table A1-2).
How many doctors in these countries are non- native English speakers?
Fourteen percent? More, or fewer? We don’t know, but we think the number is
fairly high.
IN DI A , PAK I S TA N , N IG E R I A , T HE PHIL IPPINE S ,

B A N G L A DE SH A N D E G YP T
These six large countries with old colonial ties to the UK or USA are home
to 1,900 million people, more than 25% of the world’s population. In these
Table A1-2. USA, UK, Canada, Australia—English speakers/medical

schoolsi
Native Non-native Total Medical
(millions) % (millions) % (millions) schools

USAii 256 86% 43 14% 298 178
UK iii
59 92% 5 8% 64 31
Canadaiv 19 68% 9 32% 28 17
Australia v
15 88% 2 12% 17 21
Total 349 86% 59 14% 407 247
i
Population data, same source as Table A1-1.
ii
USA “Medical schools” represent 145 accredited MD-granting institutions plus 33 accredited DO-
granting institutions; American Association of Medical Colleges, “About the AAMC,” (accessed May
12, 2016) https://www.aamc.org/about/; American Association of Colleges of Osteopathic Medicine,
“U.S. Colleges of Osteopathic Medicine,” (accessed May 12, 2016) http://www.aacom.org/become-a-
doctor/us-coms.
iii
UK General Medical Counsel, “Bodies Awarding UK Medical Degrees,” (accessed May 12,
2016) http://www.gmc-uk.org/education/undergraduate/awarding_bodies.asp.
iv
Association of Faculties of Medicine of Canada, “Accredited Canadian Medical Education
Programs,” (accessed May 12, 2016) https://www.afmc.ca/accreditation/committee-accreditation-
canadian-medical-schools-cacms/accredited-canadian-medical.
v
Australian Medical Council Ltd., “Accredited Medical Schools,” (accessed May 12, 2016) http://
www.amc.org.au/accreditation/primary-medical-education/schools.
121
Appendix 1: English Speakers Around the World 121
countries, about 422 million people—about 22%—speak English. They have

676 medical schools (Table A1-3). In these countries, getting a “good education”
includes learning English. We assume most doctors in these countries speak and
read English well.
These countries have 4.7 times as many people and 2.7 times as many medical
schools as the USA, UK, Canada and Australia combined. (Table A1-4). They have
slightly more English speakers.
E UROPE
Europe is home to about 740 million people, or about 11% of the world’s popula-
tion.2 European researchers publish widely in English-language journals. About
13% of Europeans speak English as a native language, mostly in the UK and
Ireland. Another 38%, about 191 million, speak English as an additional language.3
Table A1-3. India, Pakistan, Nigeria, The Philippines, Bangladesh, Egypt—

English speakers/medical schoolsi
English Speakers Total pop. Medical
(millions) Schools
(millions) %
Indiaii 125 10% 1,210 398
Pakistan iii
92 49% 188 97
Nigeriaiv 83 53% 156 27
Philippines v
64 64% 100 38
Bangladesh vi
30 18% 163 93
Egypt vii
28 34% 83 8
Total 422 22% 1,900 676
i
Population data, same source as Table A1-1.
ii
Medical Council of India, “List of Colleges Teaching MBBS,” (accessed May 25, 2016) http://nri-
educationalsociety.com/NRIACADEMYHOME/PDF/MCICOLLISTANDSEATS.PDF.
iii
Pakistan Medical and Dental Council, “Recognized Medical Colleges In Pakistan,” (accessed May
25, 2016) http://www.pmdc.org.pk/AboutUs/RecognizedMedicalDentalColleges/tabid/109/Default.
aspx.; Schools no longer admitting students, and recommended for closure, have been excluded.
iv
Nigeria Medical and Dental Council, “Accredited Medical Schools,” (accessed May 12,
2016) https://www.mdcn.gov.ng/page/accredited-medical-schools; only fully accredited medical
schools are counted.
v
Association of Philippine Medical Colleges, “Member Schools and Colleges,” http://www.apmcf-
ph.net/member-schools-and-colleges.
vi
Wikipedia, s.v. “List of medical colleges in Bangladesh” (accessed June 12, 2016) https://
en.wikipedia.org/wiki/List_of_medical_colleges_in_Bangladesh.
vii
Wikipedia, s.v. “List of Medical Schools in Egypt” (accessed May 12, 2016) https://en.wikipedia.
org/wiki/List_of_medical_schools_in_Egypt.
122
Table A1-4. Comparing large English-speaking countriesi
USA, UK, Canada, India, Pakistan, Ratio

Australia Nigeria, The
Philippines,
Bangladesh, Egypt
Total population 407 1900
(millions)
1: 4.7
Percent of world
5.5% 25.6%
population (%)
English speakers 407 422 1: 1.04
Medical schools 247 676 1: 2.7
i
Data from Tables A1-1 through A1-3.
English is the most widely spoken foreign language in the European Union.
Overall, about 51% of the people of the European Union speak English.4 This rep-
resents a total population four times the size of the UK. In some countries, a large
majority of the people speak English, including the Netherlands (90%), Malta
(89%), Sweden (86%), Cyprus (73%), Austria (73%) and Finland (70%).5
Based on this information, we know many European doctors read English-
language medical journals. It also seems likely their level of medical science
knowledge exceeds their level of English-language skills.
CHINA
In China, 10 million people speak English, less than 1% of the population. This
number seems poised to grow in coming years, since China has another 300 mil-
lion “learners.”6 If just 5% of these “learners” successfully learn English, it would
add a new population of English speakers the size of the UK. (A 5% rate would be
about the same as in Brazil, where relatively few people speak English.)
We assume English-language medical literacy in China is low. As more Chinese
people learn English, we expect the number of Chinese doctors who read English
language medical journals to grow even faster. Plain English medical writing
would help to speed up this process.
Conclusion
Given the large and ever-increasing number of English speakers around the world,
it makes sense to write about medical science in plain English. Many research-
ers write their articles in English, though English is not their native language.
The data suggest non-native speakers now comprise a large part of the audience
123
Appendix 1: English Speakers Around the World 123
for English-language medical journals. More doctors would read English language
journals, and understand them better, if they were written in plain English.
Notes
1. World population estimate of 7,429 million based on World Population Clock, (accessed June 13,
2016) http://www.worldometers.info/world-population/.
2. Wikipedia s.v. “Demographics of Europe,” (accessed June 12, 2016) https://en.wikipedia.org/
wiki/Demographics _of_Europe.
3. TNS Opinion & Social, Europeans and Their Languages, Special Eurobarometer 386
(Brussels: European Commission, June 2012), 5–6, 23.
4. Wikipedia, s.v. “Language in Europe,” (accessed June 21, 2016) https://en.wikipedia.org/wiki/
English_language_in _Europe.
5. TNS Opinion & Social, Europeans and Their Languages, Special Eurobarometer 386
(Brussels: European Commission, June 2012), 23.
6. Yang J, “Learners and users of English in China,” English Today 22, no. 2 (April 2006): 3–10.
124
125
Appendix 2
SELECTED EXCERPTS FROM MEDICAL

SCIENCE ARTICLES
1. Mathematical Modeling of Kidney Transport:

Glomerular Filtration1
Most mammalian kidneys have three major sections: the cortex, the outer
medulla, and the inner medulla. The outer and inner medulla are collectively
referred to as the medulla. The outer medulla may be divided into the outer stripe
and the inner stripe.
The functional unit of the kidney is the nephron; see Fig. 1. Each rat kid-
ney (which is the most well-studied mammalian kidney) is populated by about
38,000 nephrons; each human kidney consists of about a million nephrons. Each
nephron consists of an initial filtering component called the renal corpuscle and
a renal tubule specialized for reabsorption and secretion. The renal corpuscle is
composed of a glomerulus and the Bowman’s capsule. A glomerulus is a tuft of
capillaries arising from the afferent arterioles. Some of the water and solutes in
the blood supplied by the afferent arteriole are driven by a pressure gradient into
the space formed by the Bowman’s capsule. The remainder of the blood flows into
the efferent arteriole.
The most notable models of filtration of blood by glomerular capillaries are by
Deen and coworkers. Most glomerular filtration models idealize the tortuous cap-
illaries as a network of identical, parallel, rigid cylinders with homogeneous prop-
erties. Model equations typically consist of a system of coupled ODEs expressing
fluid and solute conservation:
∂ S
(QCk ) = − J k
∂x L
∂ S
(Q ) = − J v
∂x L
125
126
∂
∂x
( )
QC pr = 0
where Q denotes plasma flow rate, S and L denote the surface area and length
of the capillary, Jv and Jk denote the fluid and solute fluxes, Ck denotes the total
plasma concentration (free and bound states) of solute k, the subscript pr denotes
protein, and x denotes the position along the capillary. Boundary conditions are
given for Q, Ck, and Cpr at the afferent end of the capillary. Volume flux is assumed
to be driven by hydrostatic and oncotic pressure differences, and fluxes for small
solutes (smaller than proteins) are assumed to be both advective and diffusive,
through the fenestrated capillary walls. (wseg = 320/20.0/41.4/11.8)
2. The Frequency and Cost of Treatment Perceived

to be Futile in Critical Care2
Advances in medicine enable critical care specialists to save lives as well as prolong
dying. An admission to the intensive care unit (ICU) should be considered a thera-
peutic trial—aggressive critical care should transition to palliative care once it is
clear that the treatment will not achieve an acceptable health state for the patient.
However, intensive care interventions often sustain life under circumstances that
will not achieve an outcome that patients can meaningfully appreciate. Such treat-
ments are often perceived to be “futile” by health care providers. A survey of ICU
physicians in Canada found that as many as 87% believe that futile treatment had
been provided in their ICU in the past year. In a single-day cross-sectional study
performed in Europe, 27% of ICU clinicians believe that they provided “inappro-
priate” care to at least 1 patient, and most of the inappropriate care was deemed
such because it was excessive. (wseg = 149/24.8/37.9/14.1)
3. Regulation of GPCRs by Endocytic Membrane

Trafficking and its Potential Implications3
An essential requirement for maintenance of homeostasis in any living organ-
ism is the ability of cells to sense the external environment and, in the case of
multicellular organisms, for cells to communicate with each other via mediators
released into the extracellular milieu. Equally important are mechanisms for cells
to rapidly adapt to changes in these extracellular signals, as evident in various
disease processes characterized by uncontrolled or inadequately controlled cellu-
lar signaling. Accordingly, many important physiological processes are governed
by the coordinated actions of multiple receptor-mediated signaling pathways,
each of which is capable of rapid and specific regulation. Achieving this regulation
127
Appendix 2: Selected Excerpts 127
is highly pertinent for G protein-coupled receptors (GPCRs), which represent

the largest family of signaling receptors expressed in animals and respond to a
wide range of stimuli. The diverse physiological roles served by GPCRs, together
with evidence for disordered GPCR signaling in various pathological conditions,
emphasize the fundamental biological and clinical importance of GPCRs, and
support their prominent position as targets in drug development programs.
(wseg = 166/33.2/0.0/22.7)
Notes
1. Layton A, “Mathematical Modeling of Kidney Transport,” NIH Public Access¸ under “2.
Glomerular Filtration,” http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3745785/; originally
published in Wiley Interdiscip Rev Syst Bio Med 5, no. 5 (September 2013).
2. Huynh T, et al. “The Frequency and Cost of Treatment Perceived to be Futile in Critical Care,”
JAMA Intern Med 173, no. 20 (2013): 1888.
3. Hanyaloglu A, von Zastrow M, “Regulation of GPCRs by Endocytic Membrane Trafficking and Its
Potential Implications,” Annual Review of Pharmacology and Toxicology 48 (2008): 538.
128
â•‡ 129
Appendix 3
EXERCISE KEY
Introduction
Exercise A.â•‡ Widest reasonable audience
1. We think the widest reasonable audience includes doctors, physiologists, phar-
macists, mathematicians, biologists, bio-â•‰medical engineers, bio-â•‰chemists, and
others. It might include researchers and advanced students in these fields.
2. A mathematician or engineer probably needs more information on kidney
anatomy. A doctor probably needs a better explanation of math and fluid
dynamics. For example, what is an ODE? (An ordinary differential equa-
tion?) Why explain kidney anatomy any doctor knows, but not technical jar-
gon related to fluid dynamics some doctors don’t know (e.g., afferent, efferent,
advective, oncotic, diffusive, fenestrated)?
Exercise B.â•‡ How does sentence length affect reading ease?

1. Excerpt 1 uses the shortest sentences. Excerpt 3 uses the longest. (Table A3-â•‰1)
Table A3-â•‰1.â•‡ Average sentence length
Article Average Sentence

Length
1 Mathematical Modeling for 14
Kidney Transport
2 Futile Critical Care 24
3 Regulation of GPCR’s 35
2. We find the first excerpt, “Mathematical Modeling for Kidney Transport,” easi-
est to read. It uses shorter sentences and seems more concise than the others.
The second excerpt, on “Futile Critical Care,” uses longer sentences and seems
“wordy.” The third excerpt, “Regulation of GPCR’s by Endocytic Membrane
129
130
Trafficking,” is the hardest to read. The subject is very technical. It uses long
sentences and many long words.
3. Yes, there seems to be a general correlation between reading ease and sentence
length.
Exercise C. Recognizing symptoms of medicus incomprehensibilis

Table A3-2 identifies the symptoms of medicus incomprehensibilis.
Table A3-2. Exercise C. Symptoms of medicus incomprehensibilis observed
Excerpt 1 Excerpt 2 Excerpt 3

Low reading ease ✓ ✓ ✓
long sentence (>25 words) ✓ ✓ ✓
run-on sentence ✓ ✓
dependent clause ✓ ✓
parenthetical statement ✓
long word ✓ ✓ ✓
passive voice ✓ ✓ ✓
Abstraction ✓ ✓ ✓
abstract language ✓ ✓ ✓
nominalization ✓ ✓ ✓
formality ✓ ✓ ✓
plural subject ✓ ✓ ✓
obscure jargon ✓ ✓
Concept 1. Take charge of your reading ease score

Exercise D. Take charge of your reading ease score
1. Given the complexity of the immune system, the development
of an individual B cell is unlikely to follow a predictable and well-
executed series of decision points whereby an antigen-reactive cell
is expanded to a clone that produces a single antibody. A more real-
istic perspective is that their development depends on a series of
error-prone, random rearrangement events and mutations whereby
specificity for the original antigen is maintained (or not) by selec-
tive pressures. (wseg = 72/36.0/3.4/21.7)
2. complexity → complex; development → develop; individual → any
one; unlikely → not likely; predictable → can predict; well-executed
→ done well; decision → decide; antigen-reactive → reacts to an
131
Appendix 3: Exercise Key 131
antigen; more realistic perspective → more likely; error-prone →

prone to error; rearrangement → re-arrange; mutations → mutate;
specificity → specify; original → first; selective → select.
We consider antibody and antigen essential scientific terms.

Revision:
3. Given how complex the immune system is, any one B cell is
not likely to develop in a way that is easy to predict. For example,
a cell that reacts to an antigen expands to a clone that produces a
single antibody. More likely, the cell develops through a series of
error-prone, random events, in which it mutates and becomes re-
arranged. In the process, selective pressures determine whether the
cell still specifies the original antigen. (wseg = 75/18.7/52.4/10.6)
We find the revision easier to read than the original.
S UM M A RY
Comparing before and after wseg scores, we see reading ease improves by 49.0
points and the grade level drops by 11.1 grades (Table A3-3).
Table A3-3. Exercise D. wseg scores
Original Revision
Total Avg. Reading Grade Total Avg. words Reading Grade

words words per ease level words per sentence ease level
sentence
w s e g w s e g
Scores 72 36.0 3.4 21.7 75 18.7 52.4 10.6
Change 3 −17.3 49.0 −11.1
Chapter 1. Use normal sentence length

Exercise 1.A. Keep sentence length 15 words average, 25 words maximum
evidence-based utility for patient care, is not legally required, and
has been shown to be associated with reduced willingness for the
patient to return for future care. (wseg = 55/55.0/0.2/26.8)
132
As the wseg score shows, the original sentence has 55 words.

Revision:
ut a doctor should think twice before labeling a hospital discharge

B
as “against medical advice” and pursuing the formal process associ-
ated with it (eg, special discharge forms). Why? Because it doesn’t
help patient care, it isn’t legally required, and it makes the patient
less willing to return for future care. (wseg = 50/16.6/57.9/9.3)
The revision has three sentences with 27, 1, and 22 words. The sentences have a
total of 50 words and an average sentence length of 16.6.
2. Our Review will focus on advances in understanding of COPD and

its risk factors, prevalence, and natural history since these Reviews
were published, address some of the questions that still persist,
and raise some of the issues that health-care planners will have to
consider as the burden of COPD increases as the world’s population
ages. (wseg = 55/55.0/20.2/24.0)
Revision:
Our Review focuses on new knowledge of COPD and its risk fac-
tors, prevalence, and history since these Reviews were published.
It addresses some questions that still persist. It also raises some
issues health-care planners must think about as the burden of
COPD increases as the world’s people age. (wseg = 48/16.0/61.9/8.5)
3. Since CYP2C9 explained 6 to 10 percent of the variability in
these two patient samples, the VKORC1 genotype appears to be the
most important genetic factor determining variability in warfa-
rin dose: in both clinical populations its effect was approximately
three times that of the CYP2C9 genotype. (wseg = 46/46.0/1.9/24.4)
Revision:
CYP2C9 explains 6 to 10% of the variability in these two patient

samples. The VKORC1 genotype seems to be the key genetic factor
that determines variability in warfarin dose. In both study groups,
its effect was about three times that of the CYP2C9 genotype. (wseg
= 44/14.6/53.5/9.4)
4. A systematic review of randomized clinical trials of multiple
risk factor interventions for preventing ischaemic heart disease
had a modest effect on changes in lifestyle factors, cholesterol
concentrations, and blood pressure—the last two mainly owing to
133
the drug treatment used—but no significant effect on long term

mortality due to ischaemic heart disease. (wseg = 54/54.0/0.0/27.3)
Revision:
We reviewed randomized clinical trials of multiple risk factor inter-

ventions to prevent ischaemic heart disease. These interventions
had a modest effect on lifestyle, cholesterol levels, and blood pres-
sure. The last two were mainly owing to the drug treatment used.
Over the long term, we found no significant effect on the death rate
due to ischaemic heart disease. (wseg = 57/14.2/49.8/9.8)
5. Over the past 100 years, the science of exercise has grown from
seminal discoveries documenting the effects of exercise intensity on
vascular control, heat production, oxygen requirement, and lactic
acid dynamics—which led to Nobel Prizes in physiology or medi-
cine in 1920 (August Krogh, Denmark) and 1922 (A.V. Hill, United
Kingdom and Otto Meyerhof, Germany)—to our modern-day under-
standing that one’s cardiorespiratory fitness (indexed by one’s
maximum rate of oxygen consumption) is among the most powerful
predictors of morbidity and mortality. (wseg = 81/81.0/0.0/40.7)
Revision:
Over the past 100 years, exercise science has grown. It began with
work showing how exercise affects vascular control, heat produc-
tion, oxygen need, and lactic acid dynamics. This led to a Nobel Prize
in physiology or medicine for August Krogh of Denmark in 1920.
It led to another one for A.V. Hill of the United Kingdom and Otto
Meyerhof of Germany in 1922. Since then other studies have led to
our modern knowledge of fitness. We now know fitness, measured
by a person’s peak rate of oxygen use, strongly predicts their chance
of illness or death. (wseg = 96/16.0/53.1/9.8)
6. In a prospective study in the Netherlands that followed more
than 30,000 students 10 to 14 years of age for up to three years,
annual scoliosis screening in addition to the usual biennial health
checkup detected no cases of idiopathic scoliosis requiring surgery,
and the authors concluded that additional annual scoliosis screen-
ing was not needed. (wseg = 55/55.0/0.0/ 28.6)
Revision:
One prospective study in the Netherlands followed 30,000 stu-

dents age 10 to 14 for up to three years. Each year, a student had a
134
scoliosis screening. (This was on top of their usual health checkup

every two years.) The extra screening found no cases of idiopathic
scoliosis that required surgery. Based on this result, the authors
found the extra scoliosis screening was not needed. (wseg = 64/
12.8/51.0/9.3)
S U MM A RY
This exercise asked you to practice using normal sentence length and making
other changes to improve reading ease. Table A3-4 summarizes the wseg scores
for the original sentences and our revisions. On average, reading ease improved
by 50.8 points and grade level dropped 19.3.
Table A3-4. Exercise 1.A. wseg scores
Original Revision
Total Avg. Reading Grade Total Avg. Reading Grade

words words ease level words words per ease level
per sentence
sentence
w s e g w s e g
1 55 55.0 0.2 26.8 50 16.6 57.9 9.3

2 55 55.0 20.2 24.0 48 16.0 61.9 8.5
3 46 46.0 1.9 24.4 44 14.6 53.5 9.4
4 54 54.0 0.0 27.3 57 14.2 49.8 9.8
5 81 81.0 0.0 40.7 96 16.0 53.1 9.8
6 55 55.0 0.0 28.6 64 12.8 51.0 9.3
Average 57.7 57.7 3.7 28.6 59.8 15.0 54.4 9.4
Change 2.2 −42.6 50.8 −19.3
Exercise 1.B. Keep the subject and verb close together in the first seven or
eight words
1. Although SDM is well accepted in overtly value-laden clinical
decisions such as prostate-specific antigen testing and mammog-
raphy screening, the principles of SDM apply to a broad range of
health care decisions, discharges against medical advice included.
(wseg = 36/36.0/0.0/22.3)
Only two words separate the subject and the verb, but they do not come within
the first eight words.
135
Revision:
The values of SDM apply to a broad range of health care choices.

They are well-accepted in value-laden choices such as prostate-spe-
cific antigen testing and mammogram screening. They also apply to
a discharge against medical advice. (wseg = 36/12.0/48.9/9.4)
2. The only comprehensive effort to date to estimate summary
measures of population health for the world, by cause, is the ongo-
ing Global Burden of Diseases, Injuries, and Risk Factors (GBD)
enterprise. (wseg = 31/31.0/28.0/17.0)
Fourteen words separate the subject, effort, and the verb, is. The verb does not
come in the first eight words.
Revision:
The Global Burden of Diseases, Injuries, and Risk Factors (GBD)

enterprise is the only comprehensive effort to date to estimate, by
cause, the summary measures of population health for the world.
The GBD is ongoing. (wseg = 35/17.5/48.8/10.7)
In this revision, no words separate the subjects and verbs. The long name, “Global
Burden of Diseases, Injuries, and Risk Factors (GBD),” prevents us from putting the
subject, enterprise, within the first eight words.
3. Although drug effect is a complex phenotype that depends on

many factors, early and often dramatic examples involving succinyl-
choline and isoniazid facilitated acceptance of the fact that inheri-
tance can have an important influence on the effect of a drug. (wseg
= 39/39.0/2.3/22.6)
Only four words separate the subject and the verb, but they do not appear within
the first eight words.
Revision:
Genes have a large influence on the effect of a drug. Drug effect is a

complex phenotype that depends on many factors. Early examples
that involved succinylcholine and isoniazid helped make this fact
clear. (wseg = 34/11.3/58.4/7.9)
4. Transparent reporting of review methods and detailed reporting
of the clinical and methodological characteristics of the included
studies and their results are important to enable a reader to judge
136
the reliability of both the review and the individual studies and to
assess their relevance to clinical practice and the meaning of the
results reported in the review. (wseg = 57/57.0/0.0/28.7)
Nineteen words separate the first subject, reporting, and the verb, are. Thirteen
words separate the second subject, reporting, and the verb, are. The verb does not
come in the first eight words.
Revision:
What should a review of a primary study include? It should clearly

report review methods and include details on clinical methods. This
allows the reader to judge how reliable the review and the study
are, and to assess how relevant the study is to clinical practice. This
also helps a reader understand the results reported in the review.
(wseg = 58/14.5/55.0/9.1)
In the revision, the subjects and verbs are close together.
5. Baseline demographic characteristics and the distribution of

perioperative ARDS risk factors or modifiers among those who did
and did not develop postoperative ARDS (first procedure only) after
bleomycin therapy are presented in table 2. (wseg = 34/34.0/0.0/
23.6)
Twenty-six words separate the first subject, characteristics, and the verb, are
presented. Twenty-three words separate the second subject, distribution, and the
verb, are presented. The verb does not come in the first eight words.
Revision:
Table 2 shows demographic data and pre-and post-surgery ARDS

risk factors. It compares data for those who did and did not have
post-surgery ARDS after treatment with bleomycin. (First surgery
only.) (wseg = 33/11.0/57.2/8.0)
6. Because patients often are reluctant to discuss traumatic
events and may avoid treatment as a result, it is important to
elicit patient preferences for treatment interventions. (wseg =
26/ 26.0/21.0/16.7)
In this example, the subject and verb are together, but they don’t come within the
first eight words of the sentence.
137
Revision:
Some patients don’t want to talk about a traumatic event, and may
avoid treatment as a result. That’s why it’s important to talk with
them about their preferences for treatment. (wseg= 30/15.0/67.5/7.5)
In the revision, no words separate the subjects and verbs.
S UM M A RY
This exercise asked you to practice keeping the subject and the verb close
together in the first seven or eight words. Table A3-5 shows the wseg scores for
our revisions.
Table A3-5. Exercise 1.B. wseg scores
Original Revision

per sentence
sentence
w s e g w s e g
1 36 36.0 0.0 22.3 36 12.0 48.9 9.4
2 31 31.0 28.0 17.0 35 17.5 48.8 10.7
3 39 39.0 2.3 22.6 34 11.3 58.4 7.9
4 57 57.0 0.0 28.7 58 14.5 55.0 9.1
5 34 34.0 0.0 23.6 33 11.0 57.2 8.0
6 26 26.0 21.0 16.7 30 15.0 67.5 7.5
Average 37.2 37.2 8.6 21.8 37.7 13.6 56.0 8.8
Change 0.5 −23.6 47.4 −13.1
Exercise 1.C. Put the main point first and then give commentary,

detail or support
1. Although Ontario surgeons receive a premium and bill accurately
when they operate urgently between 12 AM and 7 AM, the times
and length of the overnight procedure and also how fatigued the
surgeon truly was when starting elective cases the next day were
unknown. (wseg = 44/44.0/16.0/21.9)
138
Seventeen words come before the main point.

Revision:
We don’t know the time or length of the overnight procedure. We

also don’t know how tired each surgeon was when they started the
elective case the next day. We do know Ontario surgeons receive a
premium when they operate urgently between 12 AM and 7 AM.
They also bill accurately. (wseg = 51/12.7/62.8/7.6)
2. First, estimates for 2000 of under-5 mortality, measured as the
probability of death between 0 years and 5 years of age (5q0), and
mortality as a young adult or middle-aged adult, measured as the
probability of death between 15 years and 60 years of age (45q15),
were developed after review of available vital registration, sample
registration, and census data and the application of the synthetic
extinct generation and growth balance methods to correct for
under-registration of deaths. (wseg = 77/77.0/0.0/36.6)
We think the main point is “estimates of … mortality… . were developed.” You

need to read the first 47 words before you can get the whole idea.
Revision:
Our first task was to estimate mortality for certain age ranges
for the year 2000. We started by reviewing data we had on births,
deaths and census. Then we corrected for under-reported deaths
using the synthetic extinct generation and growth balance methods.
We assessed under-5 mortality or 5q0, and age 15-to-60 mortality
or 45q15. We defined “under-5 mortality” as “the chance of death
between 0 and 5 years of age.” We defined “age 15 to 60 mortality”
as “the chance of death between 15 and 60 years of age.” (wseg = 89/
14.8/52.0/9.6)
3. The trial showed that pharmacogenetic- guided initiation of
warfarin therapy resulted in a greater percentage of time in the
therapeutic range, fewer excessive INRs, a shorter median time to
therapeutic INR, and fewer dose adjustments. (wseg = 34/34.0/0.6/
21.6)
This sentence already puts the main point first. But we can make it more concise,
and move the supporting details to a second sentence.
Revision:
The trial shows the benefits of starting warfarin treatment guided

by a pharmacogenetic model. They include more time in the
139
therapeutic range, fewer excess INRs, a shorter time to therapeutic

INR, and fewer dose changes. (wseg = 35/17.5/41.6/11.8)
4. The participation rate was lower than we expected when we did the
power calculations; however, taking into account the fact that more
people than expected had an increased risk and received counselling
and that not even a trend to a reduction in ischaemic heart disease
was observed, we doubt that a participation rate of 70% would have
made any difference. (wseg = 60/60.0/38.2/15.3)
We think there are two key points. One comes at the start, the other, 47 words
into the sentence. We revised to put each key point at the start of its own
sentence.
Revision:
The participation rate was lower than we planned when we did the
power calculations. But we doubt that a participation rate of 70%
would have made any difference. Though more people than expected
had increased risk and received counselling, we saw no trend towards
a decrease in ischaemic heart disease. (wseg = 50/16.6/56.2/9.5)
5. Recommended CRC screening strategies fall in 2 broad catego-
ries: stool tests that primarily detect cancer, which include detec-
tion of occult blood or exfoliated DNA, and structural tests, such
as flexible sigmoidoscopy, colonoscopy, and computed tomographic
colonography, which are effective in detecting both cancer and pre-
malignant lesions. (wseg = 46/46.0/0.0/27.4)
This sentence spreads out the main point. You need to read 27 words to get the
whole thing.
Revision:
The recommended CRC screening tests fall in 2 broad classes: stool

tests and structure tests. A stool test mainly detects cancer by
finding occult blood or fragments of DNA the body sheds in feces.
A structure test helps detect cancer or a lesion in the colon that has
not yet turned malignant. Some structure tests are: flexible sig-
moidoscopy, colonoscopy, and CT colonography. (wseg = 62/15.5/
58.7/8.9)
6. According to the AUA, the presence of three or more red blood
cells on a single, properly collected, noncontaminated urinalysis
without evidence of infection is considered clinically significant
microscopic hematuria. (wseg = 30/30.0/0.0/22.8)
140
We think the main point is broken up throughout the sentence. You don’t get the
whole main point until the end. In the revision, we broke up this long sentence to
make a short paragraph.
Revision:
When is microscopic hematuria “clinically significant?” According

to the AUA, when a single urine test shows three or more red
blood cells. The urine sample must be collected in the right way,
not contaminated, and show no sign of infection. (wseg = 40/
13.3/49.4/9.6)
S U MM A RY
This exercise asked you to practice putting the main point first before giv-
ing commentary, detail or support. Table A3-6 shows the wseg scores for our
revisions.
Table A3-6. Exercise 1.C. wseg scores
Original Revision

words words ease level words words ease level
per per
sentence sentence
w s e g w s e g
1 44 44 16 21.9 51 12.7 62.8 7.6
2 77 77 0.0 36.6 89 14.8 52.0 9.6
3 34 34 0.6 21.6 35 17.5 41.6 11.8
4 60 60 38.2 15.3 50 16.6 56.2 9.5
5 46 46 0.0 27.4 62 15.5 58.7 8.9
6 30 30 0.0 22.8 40 13.3 49.4 9.6
Average 48.5 48.5 9.1 24.3 54.5 15.1 53.5 9.5
Change 6.0 −33.4 44.3 −14.8
Chapter 2. Prefer the short word

Exercise 2.A. Keep essential scientific terms; minimize other long words
1. Therefore, it is not possible to determine whether all drug-
eluting stents could benefit from short-term regimens of dual anti-
platelet therapy based on the previous trials reported. (wseg = 26/
26.0/11.2/18.1)
141
Revision:
Therefore, there is no way to tell if a drug-eluting stent could bene-

fit from a short term course of dual anti-platelet treatment. We say
this based on the prior trials reported. (wseg = 31/15.5/62.8/8.3)
2. A careful assessment of the demographic evidence on the levels
of age-specific mortality is an integral component of any Global
Burden of Disease Study: such analyses require the sum of deaths
from specific causes to equal the independently assessed level of
mortality from all causes, for every age and sex group. (wseg = 51/
51.0/0.0/26.7)
Revision:
Any Global Burden of Disease Study must assess data with care. In
this assessment, the sum of deaths by specific cause, age and sex
must equal the sum of deaths from all causes assessed separately.
(wseg = 35/17.5/63.3/8.7)
3. Most of the pharmacogenetic traits that were first identified

were monogenic—that is, they involved only a single gene—and
most were due to genetic polymorphisms; in other words, the allele
or alleles responsible for the variation were relatively common.
(wseg = 40/40.0/34.2/13.4)
Revision:
Most pharmacogenetic traits first identified involved only one gene.

Most of these traits were due to genetic polymorphism. In other
words, the allele or alleles that caused the variation were fairly com-
mon. (wseg = 32/10.6/50.6/8.8)
4. Diagnostic accuracy is essential for good therapeutic treatment.
(wseg = 8/8.0/0.0/17.0)
Revision:
For a patient to get good treatment, the doctor needs to make the
right diagnosis. (wseg = 15/15.0/73.1/6.7)
5. The 2 objectives of the study were (1) to examine the association
between physical activity and dietary behavior and (2) to exam-
ine the potential combined effect of physical activity and dietary
behavior on biological (eg, total cholesterol) and health (eg, waist
circumference) markers. (wseg = 43/43.0/0.0/25.8)
142
Revision:
Our study checked the link between diet and exercise, and their
combined effect on bio-and health-markers (eg, total cholesterol
and waistline.) (wseg = 23/23.0/51.0/11.8)
6. Assuming vasal disruption and occlusion have been adequately
achieved during surgery, and assuming the patient adheres to using
another contraceptive method while awaiting confirmation of ste-
rility, true causes of vasectomy failure include recanalization (early
and late) and, more rarely, aberrant anatomy (e.g., the presence of a
third vas). (wseg = 48/48.0/0.0/27.7)
Revision:
Why does a vasectomy fail? The surgery may fail to disrupt and
occlude the vas. A patient may fail to use other birth control while
he waits for test results to prove he is sterile. The vas may re-cana-
lize sooner or later. Or a patient may have odd anatomy, such as a
third vas. (wseg = 54/10.8/78.3/5.0)
S U MM A RY
This exercise asked you to keep essential scientific terms but minimize other long
words. Table A3-7 shows the wseg scores for our revisions.
Original Revision

words words per ease level words words per ease level
sentence sentence
W S E G W S E G
1 26 26.0 11.2 18.1 31 15.5 62.8 8.3
2 51 51.0 0.0 26.7 35 17.5 63.3 8.7
3 40 40.0 34.2 13.4 32 10.6 50.6 8.8
4 8 8.0 0.0 17.0 15 15.0 73.1 6.7
5 43 43.0 0.0 25.8 23 23.0 51.0 11.8
6 48 48.0 0.0 27.7 54 10.8 78.3 5.0
Average 36.0 36.0 7.6 21.5 31.7 15.4 63.2 8.2
Change −4.3 −20.6 55.6 −13.2
143
Exercise 2.C. Write a compound word to promote reading ease and

show how you pronounce it
1. Predefining a threshold value for heart rate is difficult because it
must be individualized in the context of the patient’s overall hemo-
dynamic status and any preexisting comorbidities. (wseg = 27/27.0/
7.0/18.9)
Pre-defining, pre-existing, hemo-dynamic, co-morbidity.

Revision:
It is not easy to pre-define a threshold value for heart rate. Any

threshold must be set in the context of the patient’s overall blood
flow dynamics and any other pre-existing illness. (wseg = 32/16.0/
55.7/9.4)
2. For example, results of the Burden of Lung Disease (BOLD)
study—a multinational investigation of the prevalence of COPD
using a standard methodology and reported in this issue of The
Lancet, show that one of the highest prevalences of COPD was
recorded in South Africa, a country that also has a high prevalence
of tuberculosis. (wseg = 55/55.0/3.3/26.4)
Multi-national. We wouldn’t hyphenate prevalence, since it meets the three tests.

Revision:
This issue of The Lancet reports on the Burden of Lung Disease

(BOLD) study. This study checked the number of cases of COPD in
several countries using a standard method. It shows South Africa
had one of the highest numbers. South Africa also had many cases
of tuberculosis. (wseg = 48/12.0/60.7/7.7)
3. Although a recent study showed that genotype-guided dosing
led to superior control of anticoagulation, the finding was based
on a comparison with a nonrandomized, real-world parallel control
group. (wseg = 28/28.0/6.1/19.3)
Anti-coagulation and non-randomized. We wouldn’t hyphenate genotype since it

meets the three tests and qualifies as an essential scientific term.
Revision:
One new study shows dosing guided by genotype gives better con-
trol of anti-coagulation. But this finding was based on comparing
144
it to a non-randomized, real-world control group. (wseg =27/13.5/

45.8/10.2)
4. Flavonoids are naturally occurring bioactive compounds that
represent a constituent of fruits and vegetables, beyond calorie
and macronutrient content, that could potentially influence body
weight. (wseg = 25/25.0/0.0/19.6)
Bio-active, macro-nutrient.
Revision:
Flavonoids are compounds found in fruits and vegetables. They may

affect a person’s body weight. This effect goes beyond calories and
major nutrients. (wseg = 23/7.6/59.2/6.8)
5. One of the identified causes of delay in ET is multimodal imag-
ing. (wseg = 12/12.0/39.5/10.7)
Multi-modal.
Revision:
One cause of delay in ET is multi-modal imaging. (wseg = 9/9.0/

56.7/7.5)
6. The recommended modalities are photodynamic therapy, radio-
frequency ablation, or endoscopic mucosal resection. (wseg = 12/
12.0/0.0/26.4)
Photo-dynamic. Radio frequency ablation is clear without a hyphen. We wouldn’t

hyphenate endoscope, the name of a common medical instrument. Resection meets
the three tests.
Revision:
The recommended treatments are photo-dynamic therapy, radio

frequency ablation, or mucosal resection by endoscope. (wseg = 14/
14.0/0.0/20.2)
S U MM A RY
This exercise asked you to practice writing a compound word to maximize reading
ease and show how you pronounce it. Table A3-8 shows the wseg scores for our
revisions.
145
Original Revision

words Words per Ease Level Words Words per Ease Level
Sentence Sentence
w s e g w s e g
1 27 27.0 7.0 18.9 32 16.0 55.7 9.4
2 55 55.0 3.3 26.4 48 12.0 60.7 7.7
3 28 28.0 6.1 19.3 27 13.5 45.8 10.2
4 25 25.0 0.0 19.6 23 7.6 59.2 6.8
5 12 12.0 39.5 10.7 9 9.0 56.7 7.5
6 12 12.0 0.0 26.4 14 14.0 0.0 20.2
Average 26.5 26.5 9.3 20.2 25.5 12.0 46.4 10.3
Change −1.0 −14.5 37.0 −9.9
Exercise 2.D. Omit any unnecessary word ending

1. Given that this was an observational study, unmeasured con-
founding or hidden bias might exist. (wseg = 14/14.0/29.4/12.6)
We thought observational study might qualify as an essential scientific term.

Revision:
Since this was an observational study, there might be unmeasured

confounding or hidden bias. (wseg = 14/14.0/41.5/10.9)
2. However, poverty is regarded as a surrogate measure for many
factors that subsequently increase the risk of COPD, such as poor
nutritional status, crowding, exposure to pollutants including high
work exposures and high smoking rates (in countries of low and
middle income), poor access to health care, and early respiratory
infections. (wseg = 51/51.0/4.1/25.3)
We think poor nutrition or poor nutrition status works without the –al ending. You
can also say polluted air rather than pollutants. We replaced respiratory with lung.
Revision:
But being poor is a proxy for factors that raise the risk for COPD.
Poor people tend to live in crowded quarters with poor nutrition
and poor access to health care. They breathe polluted air at work.
146
In low and middle income countries, many tend to smoke. All this
helps cause early lung infection. (wseg = 54/10.8/84.6/4.1)
3. The concept of pharmacogenetics originated from the clinical
observation that there were patients with very high or very low
plasma or urinary drug concentrations, followed by the realization
that the biochemical traits leading to this variation were inherited.
(wseg = 38/8.0/0.0/23.4)
We think urine drug concentration is good modern English. The words, clinical, and
chemical are used in common speech. But in the revision, we revised to avoid using
biochemical.
Revision:
The concept of pharmacogenetics arose from clinical insight. Some

patients had a very high or a very low concentration of a drug
in their blood or urine. Doctors realized this was caused by the
patient’s genes. (wseg = 36/12.0/63.0/7.4)
4. Benefit from combined mammography and breast physical
examination screening was found in women aged 50–64, but not in
women aged 40–49. (wseg = 21/21.0/4.2/17.8)
We think exam and mammogram are good modern English. We would say,
age 50–64.
Revision:
We found a benefit from combined mammogram and breast physical

exam for women age 50 to 64. But we found none for women age 40
to 49. (wseg = 27/13.5/58.3/8.4)
5. If familial cardiomyopathy is suggested on the basis of history,
genetic testing and referral to a genetic counselor should be consid-
ered. (wseg = 21/21.0/4.2/17.8)
The ending, familial is unnecessary, since you can just say, family history of. We
think gene testing or gene counselor would sound awkward. Referral is plain English,
but refer is shorter.
Revision:
What if the patient has a family history of heart muscle disease?

In such cases, it is often best to refer them for genetic testing and
counseling. (wseg = 27/13.5/61.5/8.0)
147
6. Dosing must be tailored to the patient’s symptoms and inflam-

matory markers because up to 13% of patients required higher ini-
tial doses. (wseg = 21/21.0/32.4/13.9)
Both inflammatory markers and inflammation markers are used in scientific

presentations.
Revision:
A doctor should tailor dosing to the patient’s symptoms and inflam-

mation markers. Up to 13% of patients may need a higher initial
dose. (wseg = 23/11.5/59.0/7.8)
S UM M A RY
This exercise asked you to practice dropping any unneeded word ending. Table A3-9
shows the wseg scores for our revisions.
Table A3-9. Exercise 2.D. wseg scores
Original Revision

per per
sentence sentence
w s e g w s e g
1 14 14.0 29.4 12.6 14 14.0 41.5 10.9
2 51 51.0 4.1 25.3 54 10.8 84.6 4.1
3 38 38.0 0.0 23.4 36 12.0 63.0 7.4
4 21 21.0 4.2 17.8 27 13.5 58.3 8.4
5 21 21.0 4.2 17.8 27 13.5 61.5 8.0
6 21 21.0 32.4 13.9 23 11.5 59.0 7.8
Average 23 23.2 13 18 30 12.6 61.3 7.8
Change 7.0 −10.7 48.0 −9.9
Exercise 2.E. Avoid the noun string

1. The study by Vinden et al provides direct evidence from patient
outcomes that operating the night before is not associated with
148
increased complications for elective laparoscopic cholecystectomies

performed the following day. (wseg = 31/31.0/0.7/20.8)
Revision:
The study by Vinden et al gives proof from patient outcomes.

Operating the night before does not lead to more complications for
elective laparoscopy to remove a gall bladder the next day. (wseg =
32/16.0/55.7/9.4)
2. Performance is assessed in terms of rigorous out-of-sample pre-
dictive validity testing based on the root-mean-squared error of the
log of the age-specific death rates, the percentage of time that trend
is accurately predicted, and the coverage of the uncertainty inter-
vals (UIs). (wseg = 41/41.0/8.4/22.2)
Revision:
Through rigorous testing, we assessed how well each model per-

formed. We tested how well the model predicts by using data from
the sample. We checked the death rate for specific age groups
using the root-mean-square error of the log. We checked how
often the model accurately predicted a trend. We checked the cov-
erage of the uncertainty intervals (UIs). (wseg = 58/11.6/63.7/7.2)
We kept root-mean-square error, a standard statistical term.
3. By targeting VKOR, the post-translational modification of the

vitamin K-dependent blood-coagulation proteins is impaired.
(wseg = 14/14.0/0.0/18.5)
Revision:
Targeting VKOR impairs the synthesis of proteins that use vitamin

K to cause blood to coagulate. (wseg = 16/16.0/58.4 /9.0)
4. All participants with an unhealthy lifestyle had individually tai-
lored lifestyle counselling at all visits (at baseline and after one and
three years); those at high risk of ischaemic heart disease, accord-
ing to predefined criteria, were furthermore offered six sessions of
149
group based lifestyle counselling on smoking cessation, diet, and

physical activity. (wseg = 51/51.0/20.0/16.7)
Revision:
Each person with an unhealthy lifestyle had one-on-one counsel-

ling. The counselling took place at each visit, starting at baseline
and after one and three years. Each person with a high risk of isch-
aemic heart disease, as determined by fixed standards, was offered
six sessions of group counselling. This counselling covered how to
quit smoking, diet, and exercise. (wseg = 57/14.2/57.3/8.8)
5. As in the original CMS readmission model derivation, patients
who were discharged and then died before being rehospitalized
were not counted as “failures”; that is, they were included in the
analysis but not assigned a readmission event. (wseg = 37/37.0/
41.7/12.0)
Revision:
As with the original CMS model, we didn’t count, as a “failure,” a

patient who was discharged, but then died without ever being re-
hospitalized. We included them in the analysis, but never counted
them as re-admitted. (wseg = 36/18.0/56.0/8.4)
6. Tick paralysis, which results from gravid female bites, is a
toxin-mediated ascending paralysis that general resolves after tick
removal. (wseg = 19/19.0/4.9/17.2)
Revision:
Tick paralysis results from the bite of a pregnant female tick. Tick
paralysis is a kind of ascending paralysis caused by a toxin. It gen-
erally resolves after the tick is removed. (wseg = 31/10.3/54.4/8.2)
S UM M A RY
This exercise asked you to practice identifying and eliminating noun strings. Table
A3-10 shows the wseg scores for our revisions.
150
Table A3-10. Exercise 2.E. wseg scores
Original Revision

per per
sentence sentence
w s e g w s e g
1 31 31.0 0.7 20.8 32 16.0 55.7 9.4
2 41 41.0 8.4 22.2 58 11.6 63.7 7.2
3 14 14.0 0.0 18.5 16 16.0 58.4 9.0
4 51 51.0 20.0 16.7 57 14.2 57.3 8.8
5 37 37.0 41.7 12.0 36 18.0 56.0 8.4
6 19 19.0 4.9 17.2 31 10.3 54.4 8.2
Average 32.2 32.2 12.6 17.9 38.3 14.4 57.6 8.5
Change 6.2 −17.8 45.0 −9.4
Exercise 2.F. Don’t be afraid to start a sentence with and or but

1. Because patients with sepsis have a wide range of sympathetic
activation and responsiveness, giving a fixed dose of a β-blocker
would probably be less effective and potentially harmful if given
to all patients. Furthermore, because adrenergic stress persists as
long as the external stress (eg, infection or injury), treatment was
continued for the entire intensive care unit stay. (wseg = 58/29.0/
27.1/ 16.6)
We think “and” sounds okay. In our revision, we didn’t use any conjunction to
replace furthermore. We did use the conjunction therefore as a way to break up the
long sentences.
Revision:
A patient with sepsis can have a wide range of sympathetic activation

and response. Therefore, giving a fixed dose of a β-blocker is prob-
ably less effective and might also harm some patients. Adrenergic
stress goes on as long as the external stress from illness or injury
goes on. Therefore, we kept treating the patient for their whole ICU
stay. (wseg = 59/14.7/61.3/8.3)
2. The period since 1970 has been characterized by substantial het-
erogeneity in mortality transitions. Life expectancy in Japanese
women in 2010 was 85.9 years, and is probably higher in 2012.
151
However, the gain in the past 40 years was only 11 years, compared
with total improvements of two to three times more in other parts
of Asia (eg, the Maldives), the Middle East (especially Oman), and
Latin American (eg, Bolivia, Peru, and Guatemala). (wseg = 72/24.0/
17.9/16.7)
We think “but” sounds okay.

Revision:
Since 1970, the mean life span has changed unevenly. In 2010, a
woman in Japan could expect to live 85.9 years and likely still lon-
ger in 2012. But the gain in the past 40 years was only 11 years.
The gain was two to three times more in other parts of Asia (eg,
the Maldives), the Middle East (eg, Oman), and Latin America (eg,
Bolivia, Peru, and Guatemala). (wseg = 68/17.0/55.2/9.7)
3. Our analysis does not address the issue of whether a precise ini-
tial dose of warfarin translates into improved clinical end points,
such as a reduction in the time needed to achieve a stable thera-
peutic INR, fewer INRs that are out of range, and a reduced inci-
dence of bleeding or thromboembolic events. However, our study
lays important groundwork for a prospective trial and suggests
that such a trial should be powered to detect the benefits of incor-
porating pharmacogenetic information into the dose algorithm for
patients who require high or low doses—the subgroups in our study
for whom dose estimates based on the pharmacogenetic algorithm
differed significantly from those based on the clinical algorithm.
(wseg = 113/56.5/0.5/27.2)
These two long sentences draw a contrast between what the analysis does not
address and what it lays the groundwork for. We think just inserting the word but
would sound awkward since the contrasting ideas are so remote from each other.
We re-phrased to put these ideas closer together, and used but in the middle of
the sentence.
Revision:
Does a precise starting dose of warfarin give a better clinical result?

Does it help reduce the time needed to reach a stable therapeutic
INR? Does it mean fewer INRs out of range? Does it reduce bleeding
or thrombo-embolic events? Our study doesn’t address these issues,
but it does lay important groundwork for a future trial. The trial
should be set up to detect the benefits of using pharmacogenetic
data in the dose algorithm for a patient who needs a high or low dose.
These were the groups in our study for whom the pharmacogenetic
152
algorithm and the clinical algorithm gave significantly different

dose estimates. (wseg = 105/15.0/55.4/9.2)
4. Studies are observational in nature, prone to various biases,
and report two linked measures summarising the performance in
participants with disease (sensitivity) and without (specificity). In
addition, there is more variation between studies in the methods,
manufacturers, procedures, and outcome measurement scales used
to assess test accuracy than in randomised controlled trials, which
generally causes marked heterogeneity in results. (wseg = 59/29.5/
0.0/ 20.9)
In our revision, we used also, but not at the start of a sentence.

Revision:
These studies are observational. Therefore, they are prone to differ-

ent kinds of bias than a randomised controlled trial. For one thing,
they report two linked measures that summarize results in study
subjects, those who have the disease, and those who don’t. Thus,
they try to report on both sensitivity and specificity. The studies
also vary in their methods, manufacturers, and procedures, and
have widely differing results. They also use different scales to mea-
sure outcome and to assess test accuracy. (wseg = 79/13.1/51.0/9.4)
5. This addresses not only the challenge of selection but also sev-
eral other challenges (eg, availability and recognition). However,
participants noted several liabilities with this approach, including
lack of subspecialty specificity, additional uncompensated effort or
institutional expense, and potential for overreliance on the system.
(wseg = 43/21.5/0.0/21.6)
We think but would sound okay.

Revision:
This deals with several issues: Who should field each call? Who can
best deal with each problem? And who can best recognize each prob-
lem? But participants noted several drawbacks to this approach.
They included lack of subspecialty specificity, more effort without
more pay or overhead expense, and possible overuse of the system.
(wseg = 52/10.4/56.3/7.9)
6. This study found that older patients who had severe neurocar-
diogenic syncope (average of seven syncopal episodes in the pre-
vious two years and asystolic pauses averaging 11 seconds) had
a decreased time to first syncopal event after a pacemaker was
153
implanted. However, this study was a manufacturer-funded trial

and did not report the total syncopal burden or number of falls.
(wseg = 59/29.5/13.4/18.7)
We think but would sound okay. In our revisions, we broke up the long sentence
into shorter ones and decided not to use any conjunctions.
Revision:
This study looked at older patients who had severe fainting caused
by brief heart pauses. Study subjects had an average of seven faint-
ing spells in the prior two years. During these spells, their hearts
paused for an average of 11 seconds. After they had a pacemaker
implanted, they had less time to a first fainting spell. This study was
a manufacturer-funded trial. It did not report the total burden of
fainting or number of falls. (wseg = 75/12.5/66.6/7.0)
S UM M A RY
This exercise asked you to practice replacing a long conjunction at the beginning
of a sentence. Table A3-11 shows the wseg scores for our revisions.
Table A3-11. Exercise 2.F. wseg scores
Original Revision

per sentence
sentence
w s e g w s e g
1 58 29.0 27.1 16.6 59 14.7 61.3 8.3
2 72 24.0 17.9 16.7 68 17.0 55.2 9.7
3 113 56.5 0.5 27.2 105 15.0 55.4 9.2
4 59 29.5 0.0 20.9 79 13.1 51.0 9.4
5 43 21.5 0.0 21.6 52 10.4 56.3 7.9
6 59 29.5 13.4 18.7 75 12.5 66.6 7.0
Average 67.3 31.7 9.8 20.3 73.0 13.8 57.6 8.6
Change 5.7 −17.9 47.8 −11.7
Exercise 2.G. Avoid using a high percentage of long words

P-450 represents a deactivation pathway. For some drugs, how-
ever, oxidation leads to conversion of a prodrug into an active
154
compound. A prime example is codeine (metabolized by CYP2D6);

other examples include clopidogrel (metabolized by CYP3A4),
cyclophosphamide (metabolized by CYP2B6) and tamoxifen
(metabolized by CYP2D6).The major pathway of codeine consists
of glucuronidation and N-demethylation, whereas the CYP2D6-
mediated O-demethylation to produce morphine is a minor reac-
tion. Nevertheless, the latter is a crucial step in bioactivation,
since the affinity of codeine for the µ-opioid receptor is only 1/
200 to 1/3000 that of morphine. Previous studies have shown that
the effects of codeine—analgesic, respiratory, psychomotor, and
miotic—are markedly attenuated in people with poor metabolism
of CYP2D6. On the other hand, people with ultrarapid metabo-
lism, such as the patient described by Gasche et al. in this issue of
the Journal, produce greater amounts of morphine from codeine
and therefore may experience exaggerated pharmacologic effects
in response to regular doses of codeine. Similar effects, albeit less
dramatic, have been described in patients with ultrarapid metabo-
lism of CYP2D6 in response to routine doses of hydrocodone or
oxycodone, which are other opioids requiring CYP2D6 mediated
activation. These reports clearly illustrate the effect of CYP2D6
genetic polymorphisms on the action of codeine, ranging from vir-
tually no effect in patients with poor metabolism to severe toxic
effects in those with ultrarapid metabolism. To put these observa-
tions into perspective, these extremes of response might be rel-
evant for some 10 to 20 percent of whites who have phenotypes
associated with either poor metabolism or ultrarapid metabolism.
(wseg = 270/30.0/11.1/18.4)
We underlined 67 long words, including 13 essential scientific terms: cytochrome,

clopidogrel, cyclophosphamide, tamoxifen, glucuronidation, N- demethylation, O-
demethylation, µ-opioid, receptor, psycho-motor, hydrocodone, oxycodone, and opioids
(67 long words/270 total words = 24.8%).
Revision:
In most cases, when the body metabolizes a drug using a cyto-

chrome P-450 (“CYP-”) gene, it de-activates the drug. But in some
cases, it oxidizes a pro-drug and converts it into an active com-
pound. Examples of pro-drugs and the genes that activate them
are: codeine (CYP2D6), clopidogrel (CYP3A4), cyclophosphamide
(CYP2B6), and tamoxifen (CYP2D6).
The major pathway for codeine consists of glucuronidation and
N-demethylation. By contrast, the CYP2D6 O-demethylation to
155
produce morphine is only a minor pathway. Still, this minor path-

way is a key step to activating the drug, since a morphine molecule
is 200 to 3000 times more likely to bind with a µ-opioid receptor
than a codeine molecule.
Early studies showed that, for a person with poor CYP2D6
metabolism, codeine has much less effect on pain relief, breathing,
psycho-motor function, and pupil function. On the other hand, a
person who metabolizes ultra rapidly (such as the patient described
by Gasche et al. in this issue of the Journal) produces greater
amounts of morphine from codeine. Therefore, they may show a
greater response to regular doses.
Similar but less pronounced effects have been described in
patients with ultra rapid CYP2D6 metabolism in response to rou-
tine doses of hydrocodone or oxycodone. The CYP2D6 gene activates
these two opioids.
These reports clearly show the effect of CYP2D6 gene variation
on the action of codeine. Codeine has almost no effect in a patient
who metabolizes poorly. It has a severe toxic effect in a patient who
metabolizes ultra rapidly. These extreme effects might occur in 10–
20% of whites with a less common gene form. (wseg = 262/17.4/
44.1/11.4)
S UM M A RY
This exercise asked you to practice identifying long words and replacing or elimi-
nating them. Table A3-12 summarizes the changes.
Table A3-12. Revising to reduce non-essential long words—our results
Exercise 2.G. Original Revised % Change

Essential scientific terms 13 13 0%
Other long words 54 16 −70%
Total long words 67 29 −57%
Total words 270 262 −3%
Long words as % of total 24.8% 11.1% −55%
Flesch Reading Ease 11.1 44.1 297%
Our revision uses 29 long words and 262 total words (29 long words/262
total words = 11.1%). Overall, we reduced the number of long words by 55% and
improved the reading ease score by 30.0 points.
156
HO W DI D W E D O I T ?
We kept the 13 words we identified as essential scientific terms and used 16 other
long words. We replaced some long words with their plain-English equivalents.
Thus, analgesic becomes pain relief. Respiratory becomes breathing. Miotic becomes
pupil function.
We replaced other long words with shorter words: majority, nevertheless, affin-
ity, experience, exaggerated, similar, albeit, dramatic, illustrate, relevant, mediated,
virtually, observations, perspective, relevant, associated, markedly, and attenuated.
We also changed some nominalizations into verbs in root form. Thus, con-
version becomes converts. Metabolism becomes metabolizes. (We talk more about
nominalization in Chapter 4.)
We also broke the one paragraph into five paragraphs to reflect what we
thought was the natural progression of ideas.
Chapter 3. Omit any needless word

Exercise 3.A. Spot and omit needless words
1. In our study, we hypothesized that a heart rate range between

80/min to 94/min was a sufficient compromise between improving
cardiac performance and preserving systemic hemodynamics. (wseg
= 28/28.0/15.2/18.0)
We struck three words (3/28 = 10.7%).

Revision:
In our study, we thought a heart rate between 80 to 94/min was a

good compromise. This balances better heart performance and keep-
ing good blood flow through the system. (wseg = 30/15.0/73.1/6.7)
2. Although the definition states that this effect is in response to
noxious particles or gases, such as those in tobacco smoke, there
is also some evidence that infections can have an important role in
the presence of chronic lung inflammation in the lung. (wseg = 42/
42.0/29.2/19.6)
We struck nine words and added one for a net of eight words (8/42 = 9.0%).
Revision:
The definition states this effect is in response to noxious particles

or gases, such as those in tobacco smoke. But there is also data to
show infection plays a key role in chronic lung inflammation. (wseg
= 35/17.5/56.1/9.7)
157
3. In the multiple linear regression analysis adjusted for clinically

important covariates, four of the five common haplotypes were
found to be independently associated with the warfarin dose (P ≤
0.05) (Table 1). (wseg = 31/31.0/0.7/20.8)
We struck two words (2/31 = 6.5%).

Revision:
We processed the study data using multiple linear regression. As we

did this, we adjusted for clinically important covariates. We found
four of the five common haplotypes showed an independent link
with the warfarin dose (P ≤ 0.05). Table 1 shows the results. (wseg
= 42/10.5/49.1/9.0)
4. With the emergence of new direct acting antivirals, the treat-
ment paradigm for hepatitis C virus (HCV) infection is currently
undergoing its greatest change since the discovery of the virus was
discovered 25 years ago. (wseg = 32/32/21.0/18.2)
We struck nine words and added two, for a net of seven (7/32 = 22.9%).
Revision:
New direct acting antiviral drugs are causing a big change in the
treatment for hepatitis C virus (HCV). (wseg = 18/18.0/56.9/9.7)
5. Both dietary and exercise physical activity behavior are inde-
pendent predictors of numerous health outcomes among adults.
(wseg = 15/15.0/0.0/17.7)
We struck six words and added one, for a net of five (5/15 = 33.3%).
Revision:
Both diet and exercise predict many health outcomes in adults.

(wseg = 10/10.0/61.3/7.1)
6. For historical reasons, the American Indian/ Alaska Native
(Indian) population is particularly at risk of health and health care
disparities. We examined national data to understand how American
Indians/Alaska Natives use the health care system. To visualize the
comparison we employed an “ecology of health care” model which
uses a relative box size to indicate differences between populations.
We compared American Indians/Alaska Natives with the remain-
ing U.S. population on self-rated poor health (see accompanying
figure). This analysis reveals, as expected, the American Indian/
Alaska Native population to be significantly more rural and impov-
erished than the rest of the U.S. population non-Indians. In addition,
158
American Indians/Alaska Natives rate their health as poorer, yet

they access the health care system less often than the rest of the
U.S. population. When they do access the health care system, they
more often enter through emergency departments the ER. Despite
poorer health of American Indians/Alaska Natives, the rates of pri-
mary care visits and hospitalizations are similar to the rest of the
U.S. population non-Indians. (wseg = 168/21/22.3/15.3)
We struck 73 words and added five, for a net reduction of 68 (68/168 = 40.5%).

We expect the ecology of health care model explains itself.
Revision:
The American Indian (including the Alaska Native) suffers from

poorer health than the non-Indian, but they use health care less.
We checked US data to see how an Indian uses health care. We com-
pared them with the non-Indian on self-rated health (see figure).
The Indian is more likely poor and rural. They use health care less
but enter through the ER more. Despite poorer health, an Indian
sees a doctor or goes to the hospital about as often as a non-Indian.
(wseg = 81/13.5/63.6/7.7)
S U MM A RY
This exercise asked you to practice deleting needless words. Table A3-13 shows the
wseg scores for our revisions.
Original Revision

per sentence
sentence
w s e g w s e g
1 28 28.0 15.2 18.0 30 15.0 73.1 6.7
2 42 42.0 29.2 19.6 35 17.5 56.1 9.7
3 31 31.0 0.7 20.8 42 10.5 49.1 9.0
4 32 32.0 21.0 18.2 18 18.0 56.9 9.7
5 15 15.0 0.0 17.7 10 10.0 61.3 7.1
6 168 21.0 22.3 15.3 81 13.5 63.6 7.7
Average 52.7 28.2 14.7 18.3 36.0 14.1 60.0 8.3
Change −16.7 −14.1 45.3 −10.0
159
Exercise 3.B. Omit the needless of

1. The objective of our study was to conduct a randomized, multi-
center clinical trial to assess the effect of CPAP treatment on blood
pressure values and nocturnal blood pressure patterns of patients
with resistant hypertension and OSA. (wseg = 36/36.0/19.8/19.4)
Revision:
Our study assessed how CPAP treatment affects blood pressure in a

patient with resistant high blood pressure and OSA. This random-
ized, multi-center clinical trial assessed both day-and night-time
blood pressure. (wseg = 31/15.5/49.1/10.2)
2. Estimates of the number of deaths in children younger than
5 years from the UNPD are substantially higher than are our esti-
mations; for 2005–2010, their estimates are 8 million deaths higher
(1–6 million per year). (wseg = 35/35.0/34.3/12.8)
Revision:
The UNPD’s estimates of deaths for children under age 5 are much
higher than ours. For the years 2005 through 2010, their estimates
are 8 million higher. (This is an extra 1 to 6 million deaths per year.)
(wseg = 38/12.6/67.0/7.0)
3. Recurring themes in pharmacogenetics include the presence of a few
relatively common variant alleles of genes encoding proteins important
in drug response, a larger number of much less frequent variant alleles,
and striking differences in the types and frequencies of alleles among
different populations and ethnic groups. (wseg = 47/47.0/0.7/24.8)
Revision:
There are three common themes in pharmacogenetics. First, there

are a few common variant alleles for genes that encode proteins
important in drug response. Second, there are even more uncom-
mon variant alleles. Third, allele type and frequency differ widely
between populations and ethnic groups. (wseg = 44/11.0/49.5/9.0)
4. The quality of the studies varied considerably; many studies were
old, and few of the published studies provided sufficient detail to
replicate the intervention used. (wseg = 25/25.0/35.0/11.4)
Revision:
The quality of the studies varied a lot. Many were old. Only a few
gave enough detail so we could copy the intervention. (wseg = 23/
7.6/70.3/5.3)
160
5. At the time that the 2004 algorithm was published, there were 2
available rigorous evidence-base reviews of the treatment of RLS/
WED prepared under the auspices of the Standards of Practice
Committee of the American Academy of Sleep Medicine. (wseg = 39/
39.0/13.2/21.1)
Revision:
When the 2004 algorithm was published, two rigorous evidence-

base reviews on treating RLS/
W ED were available. These had
been prepared under the auspices of the Standards of Practice
Committee of the American Academy of Sleep Medicine. (wseg =
36/18.0/23.7/13.7)
6. Systemic symptoms (low-grade fever, fatigue, malaise, and weight
loss) occur in 30% to 50% of patients. (wseg = 16/16.0/47.8/10.5)
Revision:
Thirty to 50% of patients have systemic symptoms, such as low-grade

fever, fatigue, malaise, and weight loss. (wseg = 17/17.0/55.2/9.7)
S U MM A RY
This exercise asked you practice omitting any needless of. Table A3-14 shows the
Original Revision

sentence sentence
w s e g w s e g
1 36 36.0 19.8 19.4 31 15.5 49.1 10.2
2 35 35.0 34.3 12.8 38 12.6 67.0 7.0
3 47 47.0 0.7 24.8 44 11.0 49.5 9.0
4 25 25.0 35.0 11.4 23 7.6 70.3 5.3
5 39 39.0 13.2 21.1 36 18.0 23.7 13.7
6 16 16.0 47.5 10.0 17 17.0 55.2 9.7
Average 33.0 33.0 25.1 16.6 31.5 13.6 52.5 9.2
Change −1.5 −19.4 27.4 −7.4
161
Exercise 3.C. Omit the needless that

1. Prevalence studies estimate that 38,054 patients had a diagnosis
of a primary malignant brain tumor in the United States in 2010.
(wseg = 21/21.0/0.2/18.4)
Revision:
In 2010, about 38 thousand patients in the USA had a primary

malignant brain tumor. (wseg = 14/14.0/33.6/12.2)
2. Deaths assigned to causes that are not likely to underlie causes of
death have been reassigned with standardized algorithms. (wseg =
19/19.0/53.9/10.4)
Revision:
Deaths assigned to causes not likely to underlie causes of death have

been reassigned with standardized algorithms. (wseg = 17/17.0/
50.2/10.4)
3. Other states have simplified or eliminated special prescribing rules
(such as those requiring the use of triplicate prescription pads) that
were designed to control and monitor prescribing but that had the
(presumably unintended) effect of discouraging all prescribing of con-
trolled substances. (wseg = 41/41.0/6.3/22.5)
We think the first “that” helps, but the second does not.

Revision:
Other states have cut back on special rules on prescribing controlled

pain medicine. (For example, a rule about using a triplicate prescrip-
tion pad.) These rules tried to control and monitor prescribing pain
medicine but they instead discouraged it. (wseg = 38/12.6/53.7/8.9)
4. It has therefore been clear for some time that more effective and
tolerable treatment regimens for HCV are needed. (wseg = 19/19.0/
53.9/10.4)
Revision:
Therefore, it has long been clear: more effective and tolerable treat-

ments for HCV are needed. (wseg = 15/15.0/56.2/9.1)
5. A more recent analysis by Goyal et al revealed that both admis-
sion and post-admission hyperglycemia (admission glucose level ≤
162
3.8 mmo/L) could predict 30-day death rate in patients with AMI.

(wseg = 31/31.0/22.5/17.8)
Revision:
More recent work by Goyal et al shows high blood sugar at or after
admission predicts 30-day death rate for patients with AMI. For this
purpose, high blood sugar was defined as blood glucose ≤ 3.8 mmo/
L. (wseg = 38/19.0/69.5/8.2)
6. Overt hyperthyroidism that is inadequately treated is associated
with an increased risk of adverse maternal and neonatal outcomes
(Table 4). (wseg = 20/20.0/8.8/16.9)
Revision:
An over-active thyroid, not well treated, raises the risk of a poor

outcome for a mother and her newborn baby. See Table 4. (wseg =
23/11.5/66.4/6.8)
S U MM A RY
This exercise had you practice omitting the needless that. Table A3-15 shows our
wseg scores.
Original Revision

sentence sentence
w s e g w s e g
1 21 21.0 0.2 18.4 14 14.0 33.6 12.2
2 19 19.0 53.9 10.4 17 17.0 50.2 10.4
3 41 41.0 6.3 22.5 38 12.6 53.7 8.9
4 19 19.0 53.9 10.4 15 15.0 56.2 9.1
5 31 31.0 22.5 17.8 38 19.0 69.5 8.2
6 20 20.0 8.8 16.9 23 11.5 66.4 6.8
Average 25.2 25.2 24.3 16.1 24.2 14.9 54.9 9.3
Change −1.0 −10.3 30.7 −6.8
163
Chapter 4. Prefer active voice

Exercise 4.A. Identify active and passive voice
1. For those patients randomized to CPAP treatment, optimal CPAP
pressure was {titrated} in the sleep laboratory on a second night
by an auto CPAP device (REMstar Pro M series with C-Flex, Philips
Respironics) within a period of less than 15 days after the diagnos-
tic study to obtain a fixed CPAP pressure value, according to a previ-
ous validation by the Spanish Sleep Network.
This sentence is in passive voice.
2. A meta-analysis of data from 14 countries reported that trans-

gender female sex workers had a higher burden of HIV (27%) than
other transgender women (15%), male (15%), and female sex
workers (5%).
This sentence is in active voice.
3. Warfarin binds to albumin, and only about 3% is free and phar-
macologically active.
The first clause is active; the second is neither active nor passive.
4. After entry, the 9.6 kb viral genome undergoes cytoplasmic trans-
lation into a single polypeptide, which is subsequently {cleaved}
into 10 viral proteins—three structural and seven non-structural.
The main clause’s grammatical form looks active; but it seems passive, since
the genome isn’t acting upon anything (or upon itself). The second clause is
passive.
5. This sex difference is not clearly {understood}.
This sentence is passive.
6. Ticagrelor is {recommended} for combination therapy with aspi-

rin in patients who have acute coronary syndrome (unstable angina,
non-ST elevation myocardial infarction, or ST elevation myocardial
infarction) to reduce death from cardiovascular causes.

164
Exercise 4.B. Revise passive into active voice

1. Daily 24-hour urine collections for volume and urinary sodium
excretion were {performed} for 72 hours. (wseg = 16/16.0/26.6/13.5)

Revision:
We collected 24-hour urine samples each day for three days and
checked for volume and sodium. (wseg = 17/17.0/65.1/8.3)
2. COPD can be {classified} with respect to both phenotype and dis-
ease severity. (wseg = 12/12.0/46.6/9.7)

Revision:
We can classify a case of COPD by type and severity. (wseg = 11/

11.0/72.6/5.8)
3. Once a drug is {administered}, it is {absorbed} and {distributed}
to its site of action, where it interacts with targets (such as recep-
tors and enzymes), undergoes metabolism, and is then {excreted}.
(wseg = 31/31.0/30.7/16.6)
This sentence’s main clause is passive. Some of the other clauses are active.
Revision:
When a patient takes a drug, their body absorbs it and distributes

it to its site of action. There, it interacts with targets such as recep-
tors and enzymes, is {metabolized}, and then {excreted}. (wseg = 33/
16.5/61.9/8.7)
4. Researchers have {found} evidence for bias related to specific
design features of primary studies of diagnostic studies. (wseg = 17/
17.0/15.4/15.3)
This sentence is active.
5. The incidence of major injury in each of the cohorts was {calcu-
lated} per 10,000 person-years. (wseg = 15/15.0/11.1/15.4)

165
Revision:
For each cohort, we computed the rate of major injuries. We gave this
rate in terms of per 10,000 person years. (wseg = 21/10.5/63.2/7.0)
6. The exact pathophysiologic mechanism for scoliosis is {unknown}.
(wseg = 8/8.0/0.0/17.0)

Revision:
We don’t know exactly what causes scoliosis. (wseg = 7/7.0/42.6/ 9.0)
S UM M A RY
This exercise asked you to practice revising a passive sentence into active voice.
Table A3-16 shows the wseg scores for our revisions.
Original Revision

words words per ease level words words ease level
sentence per
sentence
w s e g w s e g
1 16 16.0 26.6 13.5 17 17.0 65.1 8.3
2 12 12.0 46.6 9.7 11 11.0 72.6 5.8
3 31 31.0 30.7 16.6 33 16.5 61.9 8.7
4 Sentence Active
5 15 15.0 11.1 15.4 21 10.5 63.2 7.0
6 8 8.0 0.0 17.0 7 7.0 42.6 9.0
Average 16.4 16.4 23.0 14.4 17.8 12.4 61.1 7.8
Change 1.4 −4.0 38.1 −6.7
Exercise 4.D. Minimize forms of to be or to have.

1. A linear regression was {used} to assess all trends over time.
(wseg = 11/11.0/72.6/5.8)
166
Revision:
We used linear regression to assess each trend over time. (wseg =

10/10.0/69.7/6.0)
2. Such assessment is not a straightforward addition of {reported}
causes. Because there are likely to be many more data {reported}
for levels of all-cause mortality than there are for individual causes,
the independent assessment of age-specific mortality is crucial to
constrain the often less robust estimates of cause-specific mortality
within each population group defined by age and sex. (wseg = 58/
29.0/14.0/18.5)
Revision:
You can’t make this assessment just by adding reported causes. Why
is this? Since there are likely more data reported for the death rate
from all causes than for individual causes. You need to assess age-
specific death rate separately. This helps double check against the
less robust estimates of death rate from specific causes for a group
defined by age and sex. (wseg = 62/12.4/65.9/7.1)
3. The current hospitalist-ambulist division of general medical care
has made important contributions to patient care, but it leaves
much to be {desired}, especially with regard to personalization and
continuity of care. (wseg = 31/31.0/3.4/20.4).
Revision:
Now days, we divide general medical care for a patient between

the doctor who works in a hospital and one who works in an office.
Sometimes, this works well. But, sometimes, it can lead to poorly
coordinated, impersonal care. (wseg = 39/13.0/61.3/7.9).
4. Although they are part of a {randomized} trial, the partici-
pants represent a {selected} group of people who have {chosen}
to participate and who attended the follow-u p. (wseg = 26/2 6.0/
30.7/15.4)
Revision:
Though part of a randomized trial, the study subjects were those

who both chose to be in the study and came for follow-up. (wseg =
23/23/69.4/9.2)
167
5. Inhaled corticosteroids (ICSs) have {had} a central role in

the management of asthma, even before publication of the first
Guidelines for the Diagnosis and Management of Asthma in 1991.
(wseg = 29/29.0/11.1/18.9)
Revision:
The guidelines for diagnosing and treating asthma first came out in
1991. Yet inhaled cortico steroids (ICSs) played a key role in treat-
ing asthma before then. (wseg = 26/13.0/50.4/9.4)
6. The cremasteric reflex, which is {elicited} by pinching the medial
thigh, causes elevation of the testicle. (wseg = 16/16.0/21.3/14.2)
Revision:
The cremasteric reflex lifts the testicle. A doctor can check this
reflex by pinching the inner thigh. (wseg = 17/8.5/63.8/6.4)
S UM M A RY
This exercise asked you to practice eliminating forms of to be and to have. Table
A3-17 shows the wseg scores for our revisions.
Original Revision

sentence sentence
w s e g w s e g
1 11 11.0 72.6 5.8 10 10.0 69.7 6.0
2 58 29.0 14.0 18.5 62 12.4 65.9 7.1
3 31 31.0 3.4 20.4 40 13.3 60.0 8.1
4 26 26.0 30.7 15.4 23 23.0 69.4 9.2
5 29 29.0 11.1 18.9 26 13.0 50.4 9.4
6 16 16.0 21.3 14.2 17 8.5 63.8 6.4
Average 28.5 23.7 25.5 15.5 29.7 13.4 63.2 7.7
Change 1.2 −10.3 37.7 −7.8
168
Exercise 4.E. Identify nominalization
1. To curb such empirical use, a report from the Infectious Diseases
Society of America (IDSA) is calling for steps to boost the devel-
opment of better diagnostic tests, to reduce regulatory hurdles
for new tests, and to improve clinical use of infectious disease
diagnostics.
2. In sub-Saharan Africa and southeast Asia, peer or community
counselling and condom distribution among female sex workers was
estimated to be cost effective, at US$86 per infection averted and
$5 per DALY averted (all costs from here expressed in 2012 US$),
and was more cost-effective than school-based education, voluntary
counselling and testing, prevention of mother-to-child transmis-
sions, and STI treatment.
3. We hypothesized that the administration of fixed-duration anti-
biotic therapy (4 days) after source control would lead to equivalent
outcomes and a shorter duration of therapy as compared with the
traditional strategy of administration of antibiotics until 2 days
after the resolution of the physiological abnormalities related
to SIRS.
4. Over 93% of participants in the control arm aged 40–49 returned
their annual questionnaire, whereas compliance with annual breast
examination screening for those in the control arm aged 50–59 var-
ied between 89% (for screen 2) and 85% (for screen 5); only ques-
tionnaires were obtained for 3% to 7% of the women.
5. For many of these reasons, evidence-based reviews generally
make authoritative statements on the degree of evidence in sup-
port of the use of each medication for a defined disorder, but they
are not always conducive to the development of practical algorithms
for the management of disorders of varying severity and a lengthy
natural history.
6. If history or examination findings raise concern for intracranial
lesions, magnetic resonance imaging of the brain can be useful for
further evaluation, with particular scrutiny of the skull base.
S U MM A RY
The purpose of this exercise was to have you identify nominalization.

169
Exercise 4.F. Convert nominalization into a verb in active voice

1. Data from each trial were entered on an intention-to-treat basis
according to the recommendations of the Cochrane Collaboration
and the Preferred Reporting Items for Systematic Reviews and
Meta-analyses (PRISMA) statement. (wseg = 30/30.0/0.0/21.6)
Revision:
We entered data from each trial on an intent-to-treat basis. We did

this as both the Cochrane Collaboration and Preferred Reporting
Items for Systematic Reviews and Meta-analyses (PRISMA) recom-
mend. (wseg = 29/14.5/25.8/13.2)
2. Accurate estimation of the number of deaths in each age and sex
group in a country, region, or worldwide is a crucial starting point
for assessment of the global burden of disease. (wseg = 32/32.0/
47.4/14.5)
Revision:
To assess the global burden of disease, start by accurately estimat-

ing deaths for each country. Break down this estimate for each age
and sex group. Then combine the data for each country to make an
estimate for the region or worldwide. (wseg = 41/13.6/67.0/7.2)
3. For some drugs, however, oxidation leads to conversion of a pro-
drug into an active compound. (wseg = 15/15.0/44.9/10.7)
Revision:
For some drugs, however, oxidizing them converts a prodrug into an

active compound. (wseg = 13/13.0/43.9/10.3)
4. After five years of counselling a significant effect on lifestyle
was seen, with a substantial reduction in the prevalence of smok-
ing, improved dietary habits, sustained physical activity (among
men), and a decrease in binge drinking. (wseg = 35/35.0/16.6/19.6)
Revision:
After five years’ counselling, we saw significant lifestyle changes: less

smoking, less binge drinking, and a better diet. Men kept up their
physical activity. (wseg = 24/12.0/50.1/9.2)
170
5. Much evidence has been amassed in support of asthma treat-

ment with ICSs. (wseg = 12/12.0/67.7/6.7)
Revision:
Much research supports treating asthma with ICSs. (wseg = 7/7.0/

66.7/5.6)
6. In its 2011 recommendation statement, the U.S. Preventative
Services Task Force did not find sufficient evidence for or against
screening for bladder cancer in asymptomatic adults. (wseg = 26/
26.0/11.2/18.1)
Revision:
In 2011, the U.S. Preventative Services Task Force found too little
evidence to recommend for or against bladder cancer screening for
an adult with no symptoms. (wseg = 26/26.0/37.2/14.5)
S U MM A RY
This exercise asked you to practice replacing nominalization with a verb in active
voice. Table A3-18 shows the wseg scores for our revisions.
Table A3-18. Exercise 4.F. wseg scores
Original Revision

per sentence
sentence
w s e g w s e g
1 30 30.0 0.0 21.6 29 14.5 25.8 13.2
2 32 32.0 47.4 14.5 41 13.6 67.0 7.2
3 15 15.0 44.9 10.7 13 13.0 43.9 10.3
4 35 35.0 16.6 19.6 24 12.0 50.1 9.2
5 12 12.0 67.7 6.7 7 7.0 66.7 5.6
6 26 26.0 11.2 18.1 26 26.0 37.2 14.5
Average 25.0 25.0 31.3 15.2 23.3 14.4 48.5 10.0
Change −1.7 −10.7 17.2 −5.2
171
Chapter 5. Prefer concrete language

Exercise 5.A. Identify whether a subject is abstract or concrete
1. Strengths of this study include the relatively large sample size,
the prospective assessment of leukocyte telomere length with blood
samples collected prior to HCT and the availability of detailed covari-
ate data known to influence transplant outcome.
The subject strengths is abstract.
2. Fourth, individuals with similar smoking and exposure histories

can vary a great deal in the severity of their disease and response to
intervention.
The subject individuals is concrete.
3. Vitamin K plays a single role in human biology—as a cofactor for

the synthesis of γ-carboxyglutamic acid.
The subject Vitamin K is concrete.
4. Previous research on systematic reviews of diagnostic tests

noted poor methods and reporting.
The subject research is abstract.
5. We defined AMI by the presence of an increase and/or decrease

of cardiac biomarkers (preferably troponin) with at least 1 value
above the 99th percentile of the upper reference limit together
with evidence of myocardial ischemia with at least 1 of the follow-
ing: (1) symptoms of ischemia, (2) electrocardiographic changes
indicative of new ischemia (new ST-T changes or new left bundle
branch block), (3) pathological Q waves on the electrocardiogram,
and (4) imaging evidence of new loss of viable myocardium or new
regional wall motion abnormality.
The subject we is concrete.
6. Documentation of the history, physical examination, diagnostic

study results, clinical impression, and diagnostic reasoning is vital
not only for medical care, but also for legal purposes.
The subject documentation is abstract.

172
Exercise 5.B. Revise abstract into concrete

1. During the past several decades, mean maternal age at delivery of a
first infant has increased steadily to 25.2 years in the United States and
30 years in Germany and Britain in 2009. (wseg = 33/33.0/22.0/18.3)
The subject mean maternal age is abstract.

Revision:
How old is a woman when she has her first baby? In 2009, the mean
age was 25.2 in the USA and 30 in Germany and Britain. The mean
age has increased steadily over the past several decades. (wseg = 38/
12.6/64.8/7.3)
2. Fifth, the airflow limitation or obstruction that happens in
COPD is caused by a mixture of small airway disease, parenchymal
destruction (emphysema), and, in many cases, increased airways
responsiveness (asthma). (wseg = 30/30.0/15.6/18.5)
The subject airflow limitation or obstruction is concrete. The word airflow sounds
real-world, but the long, Latin-origin words, limitation and obstruction sound
abstract. The words parenchymal destruction and responsiveness also sound
abstract.
Revision:
Fifth, the airflow blockage with COPD may have a few causes.
There may be a mixture of small airway disease, lung tissue damage
(emphysema), and, in many cases, other airway narrowing (asthma).
(wseg = 32/16.0/58.4/9.0)
3. The goal of therapy is to keep the INR in the therapeutic range,
since patients with an INR that is subtherapeutic are at increased risk
for thrombosis and patients with an INR that is supratherapeutic are
at increased risk for bleeding. (wseg = 41/41.0/37.2/18.2)
The subject goal is abstract.

Revision:
A doctor should try to keep the patient’s INR in the therapeutic

range. A patient with an INR below the therapeutic range has an
increased risk for thrombosis. A patient with an INR above the ther-
apeutic range has an increased risk for bleeding. (wseg = 43/14.3/
68.3/7.2)
4. Ischaemic heart disease remains a leading cause of morbidity
and mortality worldwide. (wseg = 12/12.0/25.4/12.6)
Ischaemic heart disease is a real-world problem. However, the words morbidity and
mortality sound abstract.
173
Revision:
Ischaemic heart disease remains a leading cause of illness and death

worldwide. (wseg = 12/12.0/60.7/7.7)
5. Antihypertensive medication use was retrieved from the inter-
nal pharmacy-dispensing records. (wseg = 10/10.0/0.0 /20.1)
The subject (antihypertensive medication) use is abstract.

Revision:
We took data on medicine used to treat high blood pressure from

the in-house pharmacy records. (wseg = 16/16.0/63.6/8.3)
6. Over the past 75 years, the number of U.S. women receiving pre-
natal care has steadily increased. (wseg = 16/16.0/47.8/10.5)
The subject number is abstract.

Revision:
Over the past 75 years, more and more US women have been receiv-
ing prenatal care. (wseg = 15/15.0/67.5/7.5)
S UM M A RY
This exercise asked you to tell whether a subject is abstract or concrete, and replace
an abstract subject with a concrete subject. Table A3-19 shows the wseg scores for
our revisions.
Original Revision

per sentence
sentence
w s e g w s e g
1 33 33.0 22.0 18.3 38 12.6 64.8 7.5
2 30 30.0 15.6 18.5 32 16.0 58.4 9.0
3 41 41.0 37.2 18.2 43 14.3 68.3 7.2
4 12 12.0 25.4 12.6 12 12.0 60.7 7.7
5 10 10.0 0.0 20.1 16 16.0 63.6 8.3
6 16 16.0 47.8 10.5 15 15.0 67.5 7.5
Average 23.7 23.7 24.7 16.4 26.0 14.3 63.9 7.9
Change 2.3 −9.4 39.2 −8.5
174
Exercise 5.C. Use nouns and verbs to carry the weight of meaning

evidence-based utility for patient care, is not legally required, and
has been shown to be associated with a reduced willingness for the
patient to return for future care. (wseg = 56/56.0/0.4/27.1)
We underlined 11 adjectives and adverbs (11/56 = 20%).

Revision:
But a doctor should think twice before designating a patient dis-

charge as, “against medical advice” and pursuing the formal process
associated with it (eg, special discharge form). Why? There is no evi-
dence it improves patient care. The law does not require it. Research
also shows it makes a patient less willing to return for future care.
(wseg = 56/11.2/65.5/6.8)
In the revision, we used nine adjectives and adverbs (9/56 = 16%).
2. We used these advances, and a further extension of the Brass

relational model life tables, to develop a time series of annual age-
specific mortality rates for 187 countries from 1970 to 2010, includ-
ing uncertainty. (wseg = 34/34.0/3.1/21.2)
We underlined nine adjectives (9/34 = 26%).

Revision:
We made a time series of the death rate for each year from 1970
to 2010. It covered each age group in 187 countries and included
uncertainty. We did this by using the advances we noted above,
and by extending the Brass relational model life tables. (wseg = 46/
15.3/55.1/9.3)
In the revision, we used seven adjectives (7/46 = 15%).
3. The bimodal distribution of plasma isoniazid concentrations in

subjects with genetically determined fast or slow rates of acetyla-
tion in one of those early studies strikingly illustrates the conse-
quences of inherited variations in this pathway for drug metabolism
(Fig. 2). (wseg = 39/39.0/0.0/24.1)
175
We underlined 11 adjectives and adverbs (11/39 = 28%).

Revision:
One early study showed how gene variation affects the way the body
metabolizes a drug. The study checked isoniazid in plasma for a sub-
ject with a fast or slow rate of acetylation. The results showed two
distinct modes (Fig. 2). (wseg = 40/13.3/57.9/8.4)
In the revision, we used five adjectives (5/40 = 13%).
4. There was evidence of bias when primary studies did not pro-
vide an adequate description of either the diagnostic (index) test or
the patients, when different reference tests were used for positive
and negative index tests, or when a case-control design was used.
(wseg = 42/42.0/19.1/21.0)
We underlined 10 adjectives (10/42 = 24%).

Revision:
We found some bias in the primary studies. Some did not describe
the diagnostic (index) test or the patients well enough. Sometimes,
a study used one reference test for a positive index test and another
for a negative one. Some studies used a case-control design. (wseg =
45/11.2/58.1/7.9)
In the revision, we used nine adjectives and adverbs (9/45 = 20%).
5. The use of nonergot dopamine agonists has become wide-

spread, but increasing experience with these drugs has revealed
treatment-limiting adverse effects, including the development of
augmentation and impulse control disorders. (wseg = 29/29.0/0.0/
20.5)
We underlined nine adjectives (9/29 = 31%).

Revision:
Today, a doctor often uses a non-ergot dopamine agonist to treat

a patient. But wider use has shown these types of drugs may have
a side effect that limits treatment. For example, a patient may
develop an augmentation or impulse control disorder. (wseg = 42/
14.0/53.6/9.2)
176
In the revision, we used five adjectives (5/42 = 12%).
6. Support groups could be helpful with diet maintenance. (wseg =

8/8.0/71.8/5.2)
We underlined three adjectives (3/8 = 38%).
Revision:
A support group could help a patient stick to a diet. (wseg = 11/11.0/
95.6/2.6)
In the revision, we used one adjective (1/11 = 9%).
Table A3-20. Exercise 5.C. Percentage of adjectives and adverbs
Original Revised
Adjectives & Total words Percent Adjectives & Total words Percent
adverbs adverbs
1 11 56 20% 9 56 16%
2 9 34 26% 7 46 15%
3 11 39 28% 5 40 13%
4 10 42 24% 9 45 20%
5 9 29 31% 5 42 12%
6 3 8 38% 1 11 9%
Total 53 208 25% 36 240 15%
Original Revised

per sentence
sentence
w s e g w s e g
1 56 56.0 0.4 27.1 56 11.2 65.5 6.8
2 34 34.0 3.1 21.2 46 15.3 55.1 9.3
3 39 39.0 0.0 24.1 40 13.3 57.9 8.4
4 42 42.0 19.1 21.0 45 11.2 58.1 7.9
5 29 29.0 0.0 20.5 42 14.0 53.6 9.2
6 8 8.0 71.8 5.2 11 11.0 95.6 2.6
Average 34.7 34.7 15.7 19.9 40.0 12.7 64.3 7.4
Change 5.3 −22.0 48.6 −12.5
177
S UM M A RY
This exercise asked you to practice using nouns and verbs to carry the weight of
meaning. Table A3-20 shows we reduced adjectives and adverbs to 15%.
Table A3-21 shows wseg scores for our revisions.
Exercise 5.D. Write in the singular

1. All patients had a telephone assessment of vital status and
rehospitalization at 60 and 180 days from randomization. (wseg =
18/18.0/9.9/16.3)
Revision:
We called each patient after 60 and 180 days from randomizing to

check their vital status. We also asked if they had been back in the
hospital. (wseg = 27/13.5/67.7/7.1)
2. All these hypotheses probably have elements of truth since
COPD is a classic gene- by-
environment disease with various
manifestations that include increased airways reactivity, a char-
acteristic response to infections, abnormal cellular repair, and
development of complications or comorbid disorders. (wseg = 38/
38.0/0.0/26.2)
Revision:
Each theory may have an element of truth. COPD is a classic

gene-by-environment disease. Symptoms include increased air-
way response, a poor response to infection, or poor cell repair.
A patient often has a complication or some other disease. (wseg =
38/9.5/50.2/8.6)
3. After the intake of identical doses of a given agent, some
patients may have clinically significant adverse effects,
whereas others may have no therapeutic response. (wseg = 25/
25.0/1 9.0/1 6.8)
Revision:
The same dose of a drug may cause one patient to have an adverse
effect and another to have no therapeutic response. (wseg = 22/
22.0/65.2/9.6)
4. Participants were referred to their general practitioner for medi-
cal treatment, if relevant. (wseg = 12/12.0/11.3/14.6)
178
Revision:
We referred each test subject to his or her own doctor for medical
treatment, if needed. (wseg = 16/16.0/68.9/7.6)
5. All patients had angiographically defined CAD with at least 1
vessel that met the American College of Cardiology/American Heart
Association (AHA/ACC) class I or II indications for PCI, and only
those who received implants with drug-eluting stents were consid-
ered eligible for the study. (wseg = 45/45.0/3.2/23.9)
Revision:
Each patient had CAD, as defined by heart imaging. They also had
at least one vessel that met the American College of Cardiology/
American Heart Assoc (ACC/AHA) class I or II indications for PCI.
Only those who received an implant with a drug-eluting stent took
part in the study. (wseg = 50/16.6/54.5/9.7)
6. Tinnitus occurs in most persons with normal hearing who are
exposed to silence. (wseg = 13/13.0/56.9/8.5)
Revision:
Tinnitus usually occurs in a person with normal hearing when left

in true silence. (wseg = 14/14.0/53.6/9.2)
Original Revision

per per
sentence sentence
w s e g w s e g
1 18 18.0 9.9 16.3 27 13.5 67.7 7.1
2 38 38.0 0.0 26.2 38 9.5 50.2 8.6
3 25 25.0 19.0 16.8 22 22.0 65.2 9.6
4 12 12.0 11.3 14.6 16 16.0 68.9 7.6
5 45 45.0 3.2 23.9 50 16.6 54.5 9.7
6 13 13.0 56.9 8.5 14 14.0 53.6 9.2
Average 25.2 25.2 16.7 17.7 27.8 15.3 60.0 8.6
Change 2.7 −9.9 43.3 −9.1
179
S UM M A RY
This exercise asked you to practice writing in the singular. Table A3-22 shows the
Exercise 5.E. Talk in terms of one doctor treating one patient

1. In a cohort of patients with septic shock and high risk of mortal-
ity, our open-label use of esmolol after initial hemodynamic optimi-
zation resulted in maintenance of heart rate within the target range
of 80/min to 94/min. (wseg = 38/38.0/16.8/20.3)
Revision:
Each study patient had septic shock and high risk of death. Generally,
once we stabilized a patient’s blood pressure, our open-label use of
esmolol kept their heart rate within the target range of 80 to 94/
min. (wseg = 37/18.5/60.0/9.4)
2. Use of lung function to characterize severity is, currently, the
best system available to clinicians, but it clearly falls well short of
being ideal. (wseg = 24/24.0/37.9/13.9)
Revision:
Though less than ideal, testing lung function is still the best way
for a doctor to judge how severe a patient’s COPD is. (wseg = 23/
23.0/76.8/8.2)
3. The response to many drugs in common use varies greatly among
patients. (wseg = 12/12.0/60.7/7.7)
Revision:
For many drugs in common use, the response varies greatly from
patient to patient. (wseg = 14/14.0/65.7/7.5)
4. These agents seem to facilitate the use of shortened courses of
combination interferon- free therapy, which are associated with
high (>95%) sustained response rates and relatively few toxicities.
(wseg = 27/27.0/10.2/18.5)
Revision:
These agents may allow a doctor to prescribe a shorter course of

combined treatment for a patient without using interferon. The
patient is more likely to respond well (>95%) and have fewer side
effects. (wseg = 34/17.0/57.7/9.4)
180
5. Lactate levels have become a useful marker for tissue hypoperfu-
sion and may also serve as an end point for resuscitation in patients
with sepsis and septic shock. (wseg = 27/27.0/35.2/15.0)
Revision:
A patient’s lactate levels are a useful marker for lack of blood flow
to tissue. They may also serve as an end point for reviving a patient
with sepsis or septic shock. (wseg = 32/16.0/71.6/7.2)
6. Women with hypothyroidism should be counseled about the impor-
tance of achieving euthyroidism before conception because of the risk
of decreased fertility and miscarriage. (wseg = 23/23.0/6.9/18.0)
Revision:
A doctor should talk with any woman with low thyroid function. She
needs to know that, because of the risk of low fertility or miscar-
riage, she needs to reach good thyroid function before she tries to
get pregnant. (wseg = 38/19.0/67.3/8.5)
S U MM A RY
This exercise asked you to practice talking in terms of one doctor treating one
patient. Table A3-23 shows the wseg scores for our revisions.
Table A3-23. Exercise 5.E. wseg scores
Original Revision

sentence sentence
w s e g w s e g
1 38 38.0 16.8 20.3 37 18.5 60.0 9.4
2 24 24.0 37.9 13.9 23 23 76.8 8.2
3 12 12.0 60.7 7.7 14 14.0 65.7 7.5
4 27 27.0 10.2 18.5 34 17.0 57.7 9.4
5 27 27.0 35.2 15.0 32 16.0 71.6 7.2
6 23 23.0 6.9 18.0 38 19.0 67.3 8.5
Average 25.2 25.2 28.0 15.6 29.7 17.9 66.5 8.4
Change 4.5 −7.3 38.6 −7.2
181
Chapter 6. Observe the 1066 principle

Exercise 6.A. Prefer the short word to describe the real world
1. Herpes Zoster (HV), caused by the reactivation of latent
varicella-zoster virus (VZV) manifests as an acute, painful vesicular
rash and is often accompanied by chronic pain or postherpetic neu-
ralgia. (wseg = 29/29.0/11.1/18.9)
Herpes is a real-world problem. Post-herpetic neuralgia is a kind of chronic pain.

Revision:
Herpes Zoster (HV) occurs when a latent varicella-zoster virus

(VZV) becomes active. HV shows itself as an acute, painful vesicu-
lar rash. The patient may develop post-herpetic neuralgia or other
chronic pain. (wseg = 31/10.3/38.0/10.5)
2. Fifth, the airflow limitation or obstruction that happens in
COPD is caused by a mixture of small airway disease, parenchymal
destruction (emphysema), and, in many cases, increased airways
responsiveness (asthma). (wseg = 30/30.0/15.6/18.5)
This passage talks about real-world lung damage, but the terms, limitation, obstruc-
tion, destruction, and responsiveness sound abstract. We replaced them with the
more concrete-sounding terms, blockage, damage and narrowing.
Revision:
Fifth, the airflow blockage with COPD has a few key causes. They
are small airway disease, lung tissue damage (emphysema), and,
in many cases, other airway narrowing (asthma). (wseg = 28/14.0/
56.6/8.8)
3. Maintenance of nocturnal euglycemia is extremely important
and is challenging, since most cases of severe hypoglycemia occur at
night. (wseg = 19/19.0/4.9/17.2)
Low blood sugar is a real-world problem. Measuring it is an abstract process.

Revision:
Keeping a patient’s nighttime blood sugar level near normal is both

important and challenging. Most cases of severe low blood sugar
occur at night. (wseg = 24/12.0/60.7/7.7)
182
4. Similarly, patients with diabetes had a lower risk of arte-

rial thrombosis than those without diabetes. (wseg = 15/1 5.0/
22.4/1 3.8)
The risk of a blood clot is a real-world problem. Quantifying the risk involves an
abstract math calculation.
Revision:
Likewise, a patient who had diabetes had a lower risk of arterial

thrombosis than one who did not. (wseg = 18/18.0/52.2/10.4)
5. Immune responses are orchestrated by a complex, continually
evolving cooperative network of mobile cells and their products.
(wseg = 17/17.0/15.4/15.3)
Immune response is a real-world phenomenon.

Revision:
Immune response involves an ever-changing network of mobile cells

and their products that work in concert. (wseg = 16/16.0/53.1/9.8)
6. The initial workup for urticaria and angioedema is a history and
physical examination to determine a possible etiology. (wseg = 18/
18.0/0.0/20.2)
Hives is a synonym for urticaria. Giant hives is a synonym for angioedema.1 Hives
and giant hives are real-world problems. Getting a history and doing a physical are
real-world actions. Medical diagnosis involves abstract thought.
Revision:
The initial workup for hives or giant hives is a history and physi-
cal. This data is used to figure out the likely cause. (wseg = 23/11.5/
70.1/6.3)
S U MM A RY
This exercise asked you to practice using short words to describe the real world.
Table A3-24 shows the wseg scores for our revisions.
183
Original Revision

per per
sentence sentence
w s e g w s e g
1 29 29.0 11.1 18.9 31 10.3 38.0 10.5
2 30 30.0 15.6 18.5 28 14.0 56.6 8.8
3 19 19.0 4.9 17.2 24 12.0 60.7 7.7
4 15 15.0 22.4 13.8 18 18.0 52.2 10.4
5 17 17.0 15.4 15.3 16 16.0 53.1 9.8
6 18 18.0 0.0 20.2 23 11.5 70.1 6.3
Average 21.3 21.3 11.6 17.3 23.3 13.6 55.1 8.9
Change 2.0 −7.7 43.6 −8.4
Exercise 6.B. Prefer ’s to show real-world possession or connection

1. Within each of these 2 cohorts, we compared the effectiveness
of each intervention with a control (sleep hygiene informational
video). (wseg = 20/20.0/25.7/14.6)
Cohort and effectiveness are abstract concepts. Intervention involves real-world

activity. A sleep hygiene informational video is real-world.
Revision:
Within each cohort, we compared the intervention’s effect with a

control. The control was a video on sleep hygiene. (wseg = 19/9.5/
59.1/ 7.3)
2. In this report and the accompanying appendix, we present the
data, methods and key findings of the Global Burden of Disease
Study 2010 on levels, trends, and age patterns of mortality world-
wide. (wseg = 32/32.0/28.9/17.1)
Data, method, finding, level, trend, age pattern and mortality are abstract ideas. An
appendix is part of the report.
184
Revision:
In this report, we present the data, methods and key findings from
the Global Burden of Disease Study 2010. We discuss global death
rates, trends, and age patterns. (wseg = 28/14.0/59.6/8.4)
3. Importantly, these trials all examine the initiation of therapy
with vitamin K antagonists and use as a primary end point the per-
centage of time that a patient is within the therapeutic range during
the initial phase of treatment. (wseg = 38/38.0/21.3/19.7)
This trial involves real-

world activity and abstract analysis of the results.
Therapeutic range is an abstract idea.
Revision:
These trials consider first treatment with a vitamin K antagonist.

The main study outcome is how much of the time a patient stays
within the therapeutic range during the first treatment phase.
(wseg = 32/16.0/61.0/8.7)
4. Systematic reviews of diagnostic studies involve additional chal-
lenges to those of therapeutic studies. (wseg = 13/13.0/0.0/17.6)
Reading a study is a real-world activity; analyzing it involves abstract thought.

Revision:
It is harder to do a systematic review of diagnostic studies than to

do one of treatment studies. (wseg = 18/18.0/52.2/10.4)
5. Randomized clinical trials are essential to evaluate therapies
that reduce rather than eliminate a complication of a disease.
(wseg = 18/18.0/9.9/16.3)
A disease complication might involve something you can see in the real world.
Otherwise, this sounds abstract.
Revision:
It takes a randomized clinical trial to test any treatment that

reduces a disease complication, but does not cure it. (wseg = 20/
20.0/51.5/11.0)
6. The U.S. Preventive Services Task Force recommends routine HIV
screening, known as opt-out screening, regardless of patient or phy-
sician perception of risk for all persons 15 to 65 years of age, unless
a patient refuses. (wseg = 35/35.0/31.1/17.6)
185
The risk of getting HIV is a real-world risk. What people think about that risk is
abstract.
Revision:
The US Preventive Services Task Force recommends routinely

screening each patient age 15 to 65 for HIV. This screening should be
done without regard for how the doctor or the patient perceive the
risk, unless the patient refuses. This is known as “opt-out screen-
ing.” (wseg = 44/14.6/63.1/8.0)
S UM M A RY
This exercise asked you think about real-world vs. abstract possession or connec-
tion. Table A3-25 shows the wseg scores for our revisions.
Original Revision

per sentence
sentence
w s e g w s e g
1 20 20.0 25.7 14.6 19 9.5 59.1 7.3
2 32 32.0 28.9 17.1 28 14.0 59.6 8.4
3 38 38.0 21.3 19.7 32 16.0 61.0 8.7
4 13 13.0 0.0 17.6 18 18.0 52.2 10.4
5 18 18.0 9.9 16.3 20 20.0 51.5 11.0
6 35 35.0 31.1 17.6 44 14.6 63.1 8.0
Average 26.0 26.0 19.5 17.2 26.8 15.4 57.8 9.0
Change 0.8 −10.7 38.3 −8.2
Exercise 6.C. Use terms consistently; avoid elegant variation

1. Portenoy states the problem is a lack of studies, not positive
results. Even though there is minimal literature on long-term effi-
cacy of opioids for chronic noncancer pain, the few studies that have
been published have failed to find good evidence for efficacy. (wseg =
42/21.0/42.5/12.5)
186
Revision:
Portenoy states the problem is a lack of studies, not positive results.

True, there are few studies on the long-term effect of opioids to
treat chronic non-cancer pain. But those few failed to find good evi-
dence of effectiveness. (wseg = 38/12.6/62.6/7.6)
2. Estimates from WHO’s Global Burden of Disease and Risk Factors
project show that in 2001, COPD was the fifth leading cause of death in
high-income countries, accounting for 3.8% of total deaths, and it was
the sixth leading cause of death in nations of low and middle income,
accounting for 4.9% of total deaths. (wseg = 54/54.0/23.5/23.3)
Revision:
In 2001, COPD was the fifth leading cause of death in high-income

countries, where it accounted for 3.8% of total deaths. It was also
the sixth leading cause of death in low-and middle-income coun-
tries, where it accounted for 4.9% of total deaths. (Estimates from
WHO’s Global Burden of Disease and Risk Factors project.) (wseg =
54/18.0/56.9/9.7)
3. Many legislatures and regulatory boards have adopted model pain
statutes that encourage compliance with established standards for
prescribing of pharmacologic agents for pain and other symptoms
and that protect physicians who observe these guidelines from regu-
latory intrusion and possible prosecution. (wseg = 40/40.0/0.0/25.0)
In this excerpt, we found three sets of similar terms.

Revision:
Many legislatures and regulatory boards have adopted model pain

laws. These laws encourage a doctor to comply with standards for pre-
scribing a drug to treat pain or other symptoms. They also help shield
a doctor who complies from prosecution. (wseg = 39/13.0/52.6/9.1)
4. With the emergence of new direct acting antivirals, the treatment
paradigm for hepatitis C virus (HCV) infection is currently under-
going its greatest change since the discovery of the virus 25 years
ago. New data are routinely released for different combinations
of these new agents, each reporting exceptionally high sustained
response rates for an infection that was once notoriously difficult
to treat. It is therefore difficult (even for those practicing in the
field) to keep abreast of present treatment options, or what is likely
to be available in the next 12-18 months. Because newer antivirals
187
have recently been licensed in the United States and Europe, and
the results of several promising large phase III studies have been
recently published, now is an opportune time to review the current
treatment landscape for HCV, and to anticipate how that landscape
might look in coming years. (wseg = 142/35.5/24.8/18.6)
Revision:
New direct acting anti-virals are causing the biggest change in

the treatment for hepatitis C virus (HCV) since it was discovered
25 years ago. New data is routinely released for different combina-
tions of these new anti-virals. Each report shows a high sustained
response rate. Therefore, it is hard, even for those who practice in
the field, to keep abreast of present treatment options. It is also
hard to know what will be available in the next year or so.
New anti-virals have lately been licensed in the United States
and Europe. The results of several promising large phase III studies
have now been published. Therefore, now is a good time to review
the current treatment options for HCV and to talk about how those
options might look in coming years. (wseg = 130/16.2/62.1/8.6)
5. Men taking zolpidem are at an increased risk of major injury.
Compared with the corresponding comparison cohort, the male zol-
pidem user cohort had a higher risk of major injury. (wseg = 29/
14.5/40.4/11.2)
Revision:
Men taking zolpidem had a higher risk of major injury compared to

the control group. (wseg = 15/15/56.2/9.1)
6. Celiac disease occurs in persons of European ancestry and in those
of Middle Eastern, Indian, South American, and North African descent.
It is rare in persons of Asian descent. (wseg = 29/14.5/43.3/10.8)
Revision:
Celiac disease occurs in people of European, Middle Eastern, Indian,

South American, or North African descent. It is rare in people of
Asian descent. (wseg = 24/12.0/39.5/10.7)
S UM M A RY
This exercise asked you to practice using terms consistently. Table A3-26 shows
the wseg scores for our revisions.
188
Original Revision

sentence sentence
w s e g w s e g
1 42 21.0 42.5 12.5 38 12.6 62.6 7.6
2 54 54.0 23.5 23.3 54 18.0 56.9 9.7
3 40 40.0 0.0 25.0 39 13.0 52.6 9.1
4 142 35.5 24.8 18.6 130 16.2 62.1 8.6
5 29 14.5 40.4 11.2 15 15.0 56.2 9.1
6 29 14.5 43.3 10.8 24 12.0 39.5 10.7
Average 56 29.9 29.1 16.9 50.0 14.5 55.0 9.1
Change −6.0 −15.5 25.9 −7.8
6. D. AVOI D U SI NG A L ON G , L AT IN WOR D TO DE S CR I BE

T H E R E A L WOR L D
1. The risk of HZ is elevated by 1.5 to 2 times in patients with rheu-

matic and immune-mediated diseases such as rheumatoid arthritis
(RA) and Crohn’s disease. (wseg = 26/26.0/27.5/15.8)
This sentence talks about a real-world relationship between an immune-related

disease and HZ.
Revision:
The risk of HZ is 1.5 to 2 times greater in a patient with an immune-

related disease. This would include, e.g., rheumatoid arthritis (RA)
or Crohn’s disease. (wseg = 27/13.5/58.3/8.4)
2. We hypothesize that rare ADRB2 variants modulate therapeutic
responses to LABA therapy and contribute to rare, severe adverse
events. (wseg = 19/19.0/13.8/16.0)
This sentence proposes a real-world relationship between rare ADRB2 variants

and a patient’s response to LABA treatment.
Revision:
We think rare ADRB2 variants may affect a patient’s response to

LABA treatment. They may also help cause rare but severe adverse
events. (wseg = 23/11.5/70.1/6.3)
189
3. Among the 54 patients, we identified 5 who could willfully modu-

late their brain activity (Figure 1). (wseg = 16/16.0/31.9/12.7)
This sentence talks about real-world patients who could change their brain activity.
Revision:
Among the 54 patients, we found 5 could change their brain activity

at will (Figure 1). (wseg = 16/16.0/68.9/7.6)
4. Transparency of the regulatory system is also required to over-
come several dysfunctions in the drug industry’s behaviour. (wseg =
17/17.0/0.4/17.4)
This sentence uses the word regulatory in its common, non-medical sense, which
is abstract.
Revision:
We need a regulatory system that is more open. This could help fix
some of the drug industry’s poor behavior. (wseg = 20/10.0/65.5/6.6)
5. Some experimental support exists for the concept that the abil-
ity to discriminate between “self” and “nonself” involves learning to
respond aggressively when there are signals that suggest the pres-
ence of invasive pathogens and having effective regulatory mecha-
nisms for suppressing inflammatory responses when such signals
are absent. (wseg = 46/46.0/0.0/26.2)
The working of the body’s immune response is real-world.

Revision:
To distinguish between “self” and “non-self,” the body must learn

to respond to a signal that a pathogen is present. It must also have
a way to suppress immune response when there is no such signal.
Some research supports this idea. (wseg = 41/13.6/71.2/6.7)
6. Urticaria and angioedema are thought to have similar under-
lying pathophysiological mechanisms, with histamine and other
mediators being released from mast cells and basophils. (wseg = 23/
23.0/0.0/20.0)
This sentence describes something in the real-world.

Revision:
Hives and giant hives (angioedema) have the same underlying

cause. A mast cell or a basophil gives off histamine or other immune
response. (wseg = 23/11.5/59.0/7.8)
190
S U MM A RY
This exercise asked you to practice talking about the real world using short,
concrete-sounding words. Table A3-27 shows the wseg scores for our revisions.
Original Revision

per sentence
sentence
w s e g w s e g
1 26 26.0 27.5 15.8 27 13.5 58.3 8.4
2 19 19.0 13.8 16.0 23 11.5 70.1 6.3
3 16 16.0 31.9 12.7 16 16.0 68.9 7.6
4 17 17.0 0.4 17.4 20 10.0 65.5 6.6
5 46 46.0 0.0 26.2 41 13.6 71.2 6.7
6 23 23.0 0.0 20.0 23 11.5 59.0 7.8
Average 24.5 24.5 12.3 18.0 25.0 12.7 65.5 7.2
Change 0.5 −11.8 53.2 −10.8
Chapter 7. Statistical analysis of WSEG scores

Exercise 7. Putting the tips on reading ease and vivid language into practice
Here are some of the symptoms we observed:
• Long sentence: One sentence is 39 words long, another is 36 words long. The

average sentence length (24.8 words) seems high.
• Run-on sentence: The 39-word sentence seems to cover two ideas. We would
split it into two.
• Dependent clause: The 36-word sentence starts with a long dependent clause.
This pushes the subject and verb away from the start of the sentence.
• Long words: There are many long words, but only a few essential scientific
terms. In our revision, we only used intensive care unit (ICU), therapeutic trial,
palliative care, and cross-sectional study. This paragraph has a high percentage
of long words (28/149 = 19%). This seems high for an excerpt with low science
content.
191
• Elegant variation: (a) critical care specialists, ICU physicians, ICU clinicians,
health care providers; (b) an acceptable health state for the patient, an outcome that
patients can meaningfully appreciate; (c) critical care, intensive care, aggressive
critical care, intensive care interventions, care, therapeutic, treatment; and (d) spe-
cialists, physicians, clinicians, health care providers.
• Passive voice: should be considered, are often perceived.
• Abstract language: The many long words tend to sound abstract. The narrative
uses many plural subjects and objects, which tend to sound more abstract than
those in the singular.
• Formality: The long words also tend to sound formal.
• Nominalization: admission, transition, interventions, providers.
Revision:
When a doctor treats a patient in an intensive care unit (ICU),

they can either save a life or give futile care that just delays death.
Therefore, a doctor should think of putting a patient in an ICU
as a therapeutic trial. Once the doctor sees the patient will never
recover, they should send them to palliative care.
Sadly, ICU doctors sometimes give a patient futile care. A survey
of Canadian ICU doctors found 87% believed their ICU had given
some futile care in the past year. A one-day cross-sectional study
performed in Europe found 27% of ICU doctors believed they gave at
least 1 patient “inappropriate” care. What made it so? The most com-
mon answer: it was too much. (wseg = 120/15.0/61.8/8.3)
Chapter 8. Organize your narrative in a way

that’s helpful for your reader
Exercise 8.A. Introduce and develop a single idea in each paragraph
1. We find this paragraph challenging and we feel we need to study it. We think it
covers multiple ideas.
Revision:
What happens during metabolism?

(31 words)
192
What are some examples of a CYP-gene activating a pro-drug?

ples include clopidogrel (metabolized by CYP3A4), cyclophospha-
mide (metabolized by CYP2B6) and tamoxifen (metabolized by
CYP2D6). (24 words)
What are the metabolic processes CYP2D6 triggers in codeine?
The major pathway of codeine consists of glucuronidation and N-
demethylation, whereas the CYP2D6-mediated O-demethylation to
produce morphine is a minor reaction. Nevertheless, the latter is a
crucial step in bioactivation, since the affinity of codeine for the µ-
opioid receptor is only 1/200 to 1/3000 that of morphine. (47 words)
How does genetic variation effect how a patient metabolizes codeine?
Previous studies have shown that the effects of codeine—analgesic,
respiratory, psychomotor, and miotic—are markedly attenuated in
people with poor metabolism of CYP2D6. On the other hand, peo-
ple with ultrarapid metabolism, such as the patient described by
Gasche et al. in this issue of the Journal, produce greater amounts
of morphine from codeine and therefore may experience exagger-
ated pharmacologic effects in response to regular doses of codeine.
(68 words)
How does genetic variation effect how a patient metabolizes other drugs?
Similar effects, albeit less dramatic, have been described in patients
with ultrarapid metabolism of CYP2D6 in response to routine doses
of hydrocodone or oxycodone, which are other opioids requiring
CYP2D6 mediated activation. These reports clearly illustrate the
effect of CYP2D6 genetic polymorphisms on the action of codeine,
ranging from virtually no effect in patients with poor metabolism to
severe toxic effects in those with ultrarapid metabolism. (66 words)
How common is it for a patient to have a genetic variation that effects
how they metabolize a drug?
To put these observations into perspective, these extremes of
response might be relevant for some 10 to 20 percent of whites who
have phenotypes associated with either poor metabolism or ultra
rapid metabolism. (32 words)
We think the shorter paragraphs and headings help a reader to scan the article
and get the “big picture” quickly before reading the details.
2. We feel we need to study this paragraph. We think this paragraph covers

multiple ideas.
193
Revision:
Pulmonary Artery Catheter (PAC)—

The clinical gold standard for
estimating CO
The [pulmonary artery catheter] PAC, today still considered the
clinical gold standard for CO estimation, gives three fundamental
bits of hemodynamic information: CO, pulmonary and cardiac fill-
ing pressures and mixed SvO2. The PAC is considered the clinical
gold standard or reference method for CO estimation and every new
tool built for CO estimation has to be compared with PAC in valida-
tion studies. (59 words)
How can you use the PAC to measure macro circulation?
The technique is based on the injection of an ice-cold solution into
the right atrium (proximal port—prox injectate—blue lumen). The
change in blood temperature is measured in the pulmonary artery by
a thermistor placed proximally to the tip of the catheter. The thermo
dilution curve is used for CO estimation by means of the Steward-
Hamilton formulation. The measurement is repeated at least three
to five times (in order to compensate for variations induced by the
respiratory cycle) and the average value is then calculated. (86 words)
How can you use a PAC to obtain an inverted thermodilution curve?
An “inverted” thermodilution curve can be obtained by worm-
ing [sic] the blood with a thermal filament (VigilanceTM, Edwards
Lifesciences, Irvine, CA, USA) or a thermal coil (OptiQTM, ICU
Medical, San Clemente, CA, USA). This system allows for a semi-
continuous CO measurement by displaying average CO Values
for the previous 10 min and limiting, for this reason, the effects
of arrhythmias and other compounding factors. Although the CO
value is frequently updated, this type of monitoring is not continu-
ous (beat-by-beat) and is therefore less accurate and rapid in detect-
ing hemodynamic instability. (89 words)
What are limiting factors for estimating CO?
Major limiting factors for CO estimation with any type of thermodi-
lution is the occurrence of tricuspid regurgitation and intracardiac
shunts. In fact, prolonged indicator transit times or indicator recy-
cling may lead to errors in CO estimation. SvO2 suggests whether
or not cardiac output is adequate in a patient since it provides a
useful indication about the adequacy of tissue oxygenation in spe-
cific conditions of metabolic activity. When SvO2 decreases below
normal values (70–75%) in the presence of normal arterial oxygen
saturation and without anemia, it means that CO is inadequate
194
and measures aimed at increasing DO2 should be promptly imple-

mented. (100 words)
What are the different ways to measure SvO2?
SvO2 can be measured either continuously (fiberoptic fibers) or
intermittently by withdrawing a mixed venous blood sample from
the distal lumen of the PAC. The PAC measures: the pulmonary
artery pressure which represents right ventricular afterload; right
atrial pressure (RAP/CVP, PAOP) have been demonstrated to be
less reliable than dynamic indicators of fluid responsiveness (pulse
pressure variations PPV, systolic pressure variation SPV and stroke
volume variation SVV). Nonetheless, when static parameters are
particularly low, they can be considered to be as reliable as dynamic
ones. (86 words)
What is the recent trend in using PAC?
Over recent years, the use of PAC has decreased significantly for
two main reasons: firstly, several randomized and non-randomized
studies have not demonstrated an improvement in patient out-
comes when therapies were guided by PAC and secondly nowadays
the PAC is the most invasive tool for hemodynamic monitoring cur-
rently used in ICUs or ORs. In 1996, Connors and co-workers pub-
lished in JAMA a prospective cohort study in which the authors used
case-matching multivariable regression modeling techniques and a
propensity score. The results showed that pulmonary artery cath-
eterization in critically ill patients was associated with an increased
risk of death (odds ratio 1.24; 95% confidence interval 1.03–1.49)
as well as a prolonged length of stay and increased resource utiliza-
tion. (117 words)
What was the aftermath of the Connors and co-workers study?
After this shocking paper, a litany of randomized trials aimed at con-
firming these results was performed in ICUs. The net results of this
great amount of data corresponded to an increased suspicion that
PAC-guided treatment may not be superior to non-PAC-guided treat-
ment and a marked decrease in PAC use worldwide has become clear
during the last 10–15 years. In the USA, Wiener et al. reported a 65%
decrease in PAC use between 1993 and 2004 and in Canada, Koo et al.
reported a more than 50% decrease between 2002 and 2006. A great
number of papers has been published by opinion leaders and general
considerations about the use of PAC and its benefits. (113 words)
What are the doubts about the usefulness of PAC?
The doubts regarding its usefulness can be listed as follows.
(10 words)
195
We think the shorter paragraphs and headings help make the content easier
to grasp.
Exercise 8.B. Present two-dimensional data in a table, chart or graph

1. All 525 study participants, who were randomized to receive var-
enicline or placebo, had been diagnosed with major depressive dis-
order and were being treated with antidepressant drugs at a stable
dose or had been successfully treated for depression within the past
2 years. At 9 to 12 weeks, 35.9% of those who received varenicline
quit vs 15.6% taking placebo; at 40 weeks, 20.3% of the varenicline
group had quit compared with 10.4% of the placebo group. Depression
and anxiety did not increase in either group, but the researchers cau-
tioned that their findings may not apply to smokers whose depres-
sion isn’t successfully treated. (wseg = 101/33.6/30.4/15.2)
Revision:
Each of the 525 study participants was chosen at random to

receive varenicline or placebo. Each had been diagnosed with
major depressive disorder. Each was being treated with anti-
depressant drugs at a stable dose or had been treated for depres-
sion, with success, within the past 2 years. Table A3-28 shows the
results.
Table A3-28. Study subjects who quit smoking
Received varenicline Received placebo

At 9 to 12 weeks 35.9% 15.6%
At 40 weeks 20.3% 10.4%
Depression and anxiety did not increase in either group. But the
researchers cautioned, their findings might not apply to a smoker
whose depression is not treated with success. (wseg = 80/13.5/55.8/
8.7, excludes table)
2. What about health in Scotland? According to the UK’s national
statistical office, healthy life expectancy was 59·8 years for men
and 64·1 years for women in Scotland during 2008–10, 4·6 years
and 2·3 years fewer than for men and women in England, respec-
tively. According to the British Heart Foundation, 35% of Scottish
men and 30% of women have high blood pressure; alcohol use is one
noticeable contributor to ill health in Scotland, with up to 50% of
196
men and 30% of women exceeding guidelines for drinking. (wseg =

87/29.0/37.9/13.3)
Revision:
What about health in Scotland? (Table A3-29)
Table A3-29. Some key data on health in Scotland
Men Women
Healthy life expectancy (years)1 59.8 64.1
Difference compared to England (years) 1
(4.6) (2.3)
High blood pressure2 35% 30%
Alcohol use exceeding guidelines 2
50% 30%
Sources:
1. UK national statistical office; data for 2008–10
2. British Heart Foundation
Alcohol use is one noticeable contributor to ill health in Scotland.
(wseg = 18/8.0/55.9/7.4, excludes table)

You might instead present the data in the form of a chart or graph. Figures A3-1
and A3-2.
70
60
50
40
30
20
10
0
Men Women
England Scotland
Figure A3-1 Life Expectancy in Years.1

197
% with Alcohol use exceeding

guidelines
Women
Men
% with High Blood Pressure
0 10 20 30 40 50 60
Figure A3-2 Scotland Health. 2
Sources: 1. UK national statistical office; data for 2008–10

2. British Heart Foundation.
Chapter 9. Choose a clear narrative pathway

Exercise 9.A. Start with things known
We don’t think the excerpt starts by talking about concepts familiar to the widest
reasonable audience. It starts by reviewing kidney anatomy, which an engineer or
a mathematician might not know.
Exercise 9.B. Start by anchoring the discussion in the real world

You might say the narrative starts by talking about the real world since a kidney
is something in the real world. But “mammal” is the name for an abstract bio-
logical class. Likewise, the concept of a “mammal kidney” is an abstract idea. It
assumes kidneys from different mammals are enough alike that it makes sense
to generalize.
Exercise 9.C. Choose a good narrative pathway

1. We see three narrative pathways: (a) kidney anatomy—big to small, (b) rat
kidney vs. human kidney, and (c) blood flow—upstream to downstream.
2. The narrative doesn’t state them explicitly, but they are clear from context.
3. The narrative doesn’t follow them consistently. We saw some issues that inter-
rupt the narrative flow:
• The narrative of “kidney anatomy— big to small” changes direction
at one point and goes from smaller to bigger (inner medulla + outer
medulla = medulla).
198
• The big-to-small narrative also gets interrupted by comparing a rat kidney

to a human kidney.
• There is a gap in the big-to-small narrative. It never explains where a
nephron belongs in relation to the large parts of the kidney: cortex, inner
medulla, and outer medulla.
Exercise 9.D. Make a smooth transition between concrete and abstract

1. The real world objects the narrative mentions are cortex, inner and outer
medulla, nephron, rat kidney, human kidney, renal corpuscle, renal tubule,
glomerulus, Bowman’s capsule, capillaries, arterioles, water and blood. The main
real world action is blood flow.
The concept of a mammal is an abstract biological classification. Likewise, the
generalized concept of a mammal kidney is abstract. The ideas, “specialized for
absorption and secretion,” and “pressure gradient” also sound abstract.
It might be possible to revise the discussion of real-world blood flow to use
more concrete-sounding terms (e.g., soaks up or absorbs rather than absorp-
tion; and empties into, gives off, or secretes rather than secretion.) It might also
make for a smoother transition to keep real-world anatomy, real-world blood
flow, and abstract ideas in separate sentences or paragraphs.
2. Both a rat kidney and a human kidney are made up of bundles of tiny neph-
rons that filter blood. The size and shape of an individual nephron are similar.
The main difference is, a human kidney is bigger because it has many more
nephrons. The narrative doesn’t mention this.
3. All mammal kidneys consist of bundles of tiny nephrons. As the size of the
mammal increases, the size and shape of an individual nephron stays about
the same, but the number of nephrons increases. The narrative also doesn’t
mention this.
4. All mammal kidneys are enough alike that it makes sense to generalize about
a typical mammal kidney. A kidney, whether of a rat, a human, an elephant or
a whale, is composed of bundles of tiny nephrons that filter wastes and toxins
out of the blood and channels them to make urine. The size and shape of an
individual nephron is similar for any mammal. Bigger mammals have more
nephrons: a rat kidney has about 38,000 nephrons. A human kidney has about
1 million nephrons.
Chapter 10. Forge a strong chain of logical

reasoning
Exercise 10.A. Explain each step of reasoning

1. A nephron is located partly in the cortex and partly in the medulla.
199
2. The narrative doesn’t explain this. Therefore, it appears to have left out a step
of reasoning. (The article does show this later in a figure.)
Exercise 10.B. State the problem before you solve it

1. If a patient’s kidney stops filtering blood, they will die unless they receive
dialysis.
2. A math model of how the kidney filters blood might help design a dialysis
machine that better mimics kidney function. It might also lead to better treat-
ment or prevention of kidney disease.
3. There is no statement of the problem math modeling helps solve in this
excerpt. (But the article does mention this elsewhere.)
Exercise 10.C. Say it in words before you say it in symbols

1. The narrative explains the three equations in general terms but doesn’t explain
each individual equation. We found the narrative confusing. We would like to
see the equations explained more clearly. For example, the narrative describes
the three equations as ODE’s (ordinary differential equations?) but uses the
notation for partial derivatives (δx). We also found the explanation of variables
unclear (e.g., what does Cpr stand for?)
2. The widest reasonable audience probably needs: (1) for those who have for-
gotten calculus, a brief summary of the model; and (2) for those who haven’t
forgotten calculus, a careful, step-by-step tour.
Note
1.
Stedman’s Medical Dictionary, s.vv. “Urticaria,” “Angioedema.”
200
201
GLOSSARY
The 1066 Principle—the general tendency for English speakers to use short words to talk about the
real world, and long words, more sparingly, to talk about abstract ideas.
Abstract—a theoretical way of looking at things; something that exists only in idealized form.
A term is abstract if it relates to the world of ideas, including a concept, theory, calculation or
procedure. Contrast with concrete.
Active voice—A sentence is in active voice when its subject is doing the action. See voice.
Clear—Writing is clear when the narrative uses words and concepts familiar to the reader. Ideally,
a reader can understand and vividly imagine the article on first reading without having to study it.
The reader remembers each key idea.
Closed compound—a compound word written as one word (e.g., multicell, hyperadrenergic, vasomotor,
and sinoatrial). See compound word.
Compound word or compound—a word formed by combining two or more words, or a word plus a
prefix. The three main types of compounds are the open compound (e.g., student nurse), the closed
compound (e.g., multicell), and the hyphenated compound (e.g., pre-menstrual).
Concise—Writing is concise when it demands as little of the reader’s mental energy as possible.
This usually means short while still clear. Good writing involves tradeoffs. A few short words may
convey the message more vividly than one long, but lifeless word. Writing concisely often means
cutting any unnecessary word; but sometimes, cutting too many words makes the message cryptic
and harder to understand.
Concrete—something from the real world (e.g., a doctor, a patient, a bed, a test tube). Contrast with
abstract.
Elegant variation—varying terms to make writing more interesting. Technical writing tends to
avoid elegant variation, but it is common in other types of writing.
Essential scientific content—important scientific ideas an author must include in their article.
Essential scientific term—a long word that helps convey essential scientific content clearly and
concisely. An essential scientific term meets four tests:
1. No shorter word serves just as well,
2. You can’t paraphrase in a few short words,
3. Doctors and other medical scientists use the term consistently (i.e., exclusively), and
4. It’s easy to look up in a standard reference.
False signal—using an abstract-sounding word to talk about something in the real world, or a
concrete-sounding word to talk about something abstract. Contrast with signal.
201
202
202 Glossary
Flesch Reading Ease—a readability test that indicates how difficult it is to read a passage in English.
The scores generally range from 0.0 to 100.0.
Flesch-Kincaid Grade Level—a readability test that assigns a USA school grade level or year to a
passage in English.
Hyphenated compound—a compound word where words are written together but separated by a
hyphen (e.g., pre-menstrual, cost-effective, one-time, self-reported). See compound word.
Insider—somebody who knows the science and vocabulary of a particular specialized field. Insiders
are the narrowest possible definition of the potential audience for an article. Contrast with the
widest reasonable audience.
Long word—any word with three or more syllables, but not including a two-syllable word that
becomes a three-syllable word by adding a common ending, such as -ed, -es or -ing.
Medicus incomprehensibilis—a condition that affects doctors and other medical scientists and causes
them to write dull, lifeless prose that is hard to understand. Medicus incomprehensibilis is primarily
caused by needless grammatical complexity.
Narrative pathway—the direction of a narrative, or a conceptual program for organizing a narrative.
Nominalization—the process of making an abstract noun out of a verb or adjective.
Noun string—a group of nouns and their modifiers. Often, a noun string consists of obscure
technical terms. Multiple terms may function together as an adjective.
Open compound—a compound word, where words work together, but are written as separate words
(e.g., student nurse, 50 percent, reference book). See compound word.
Passive voice—A sentence is in passive voice when its subject receives the action. See voice.
Past participle—a form of the verb that expresses completed action.1 A past participle is usually the
same form as the verb in past tense. For regular verbs, this means adding a -d or -ed ending (e.g.,
worked, decided, starved). Irregular verbs use irregular forms (e.g., broken, swum).2 Examples:
• “The results of the meta-analysis of treatment effect of lubiprostone vs. placebo are shown in
Figures 2 and 3.”3 In this sentence, shown is a past participle.
• “It may be specified in the protocol of a prospective accuracy study, for instance, that to reduce
study costs or burden to patients only a randomly selected subset of patients in a specific
subgroup are to be verified by the preferred reference standard.”4 In this sentence, specified,
selected, verified, and preferred are past participles.
Plain English—writing that conveys the right content, clearly and concisely. Writing in plain
English involves sharpening up the medical science to make it clearer and more accessible to the
widest reasonable audience.
Short word—a one-or two-syllable word. This also includes a two-syllable word that becomes a
three-syllable word by adding a common ending, such as -ed, -es, or -ing. Contrast with long word.
Signal—a way of indicating, through word choice, whether you’re talking about the real world
or the world of abstract ideas. Short words tend to signal real world, and longer words abstract.
Contrast with false signal.
Subject (grammar)—the noun or pronoun that agrees with the verb.5 A noun functioning as a
subject is the actor, person, or thing about which an assertion is made in a clause.6 Examples:
• “Onychomycosis is a fungal infection of the nails that causes discoloration, thickening, and
separation from the nail bed.”7 In this sentence, Onychomycosis is the subject.
• “Identification of hyphae, pseudohyphae, or spores confirms infection but does not identify the
organism.” 8 Here, Identification is the subject. (The phrase, “Identification of hyphae, pseudohyphae,
or spores,” is the logical subject.9)
203
Glossary 203
Verb—expresses an action, occurrence or a state of being.10 Examples:

• “Diagnostic studies typically evaluate the accuracy of one or more tests, markers, or models by
comparing their results with those of, ideally, a “gold” reference test or standard.”11 Here, the
verb is evaluate.
• “High K+ intake also has a stimulatory effect on the release of aldosterone at the level of the
adrenal gland.”12 Here, the verb is has.
Vivid language—language that is clear, detailed, powerful, full of life, and strikingly alive.
Voice—a term that describes whether the subject of the sentence is doing or receiving the action.
See active voice; passive voice.
Widest reasonable audience—the widest reasonable audience for a journal article includes anybody
with an interest in the science, whether or not they are an insider in the field. It includes a doctor
or scientist working in the same specialty, another specialty, or even another discipline. It includes
someone living or educated in an English-speaking country or elsewhere, whether they are a native
speaker of English or not. It includes readers at different levels of training. It includes a regular
journal subscriber and somebody who searches for an article on the internet. Contrast with insider.
WSEG—four items of data helpful for assessing reading ease for a writing sample: the number of
words (w), average sentence length (s), Flesch Reading Ease score (e), and Flesch-Kincaid Grade
Level (g).We write a wseg score for a writing sample in the form (wseg = 55/55.0/0.2/26.8).
Notes
1. Chicago Manual of Style, 15th ed. §5.103.
3. Li F, et al. “Lubiprostone is Effective in the Treatment of Chronic Idiopathic Constipation and
Irritable Bowel Syndrome: A Systematic Review and Meta-Analysis of Randomized Controlled
Trials,” Mayo Clinic Proc 91, no. 4 (2016): 461.
4. Naaktgeboren C, et al. “Anticipating Missing Reference Standard Data When Planning
Diagnostic Accuracy Studies,” BMJ 352 (2016), under “The problem: missing reference stan-
dard data,” http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4772780/.
5. Cutts, Oxford Guide to Plain English, 122 (see chap. 1, n. 6).
6. The Chicago Manual of Style, 15th ed. §5.23.
7. Westerberg D, Voyack M, “Onychomycosis: Current Trends in Diagnosis and Treatment,” Am
Fam Phys 88, no. 11 (2013): 762.
8. Ibid.
9. See Williams, Style: Lessons in Clarity, 81–82 (see Preface, n. 8).
10. Merriam-Webster’s Learner’s Dictionary, Merriam- Webster.com, (accessed July 11, 2016),
http://www.merriam-webster.com/dictionary/verb. s.v. “Verb.”
11. Naaktgeboren et al. “Anticipating Missing Reference Standard,” (see Glossary, n. 4).
12. Palmer B, Clegg D, “Achieving the Benefits of High-Potassium, Paleolithic Diet, Without the
Toxicity,” Mayo Clinic Proc 91, no. 4 (2016): 500.
204
205
RESOURCES
Books
Booth, Wayne C., Gregory G. Colomb, Joseph M. Williams. The Craft of Research. 3rd ed. Chicago:
University of Chicago Press, 2008.
Cutts, Martin. Oxford Guide to Plain English. Oxford: Oxford University Press, 2009.
Garner, Bryan A. The Elements of Legal Style. 2nd ed. Oxford: Oxford University Press, 2002.
Garner, Bryan A. Legal Writing in Plain English. Chicago: University of Chicago Press, 2001.
Garner, Bryan A. Securities Disclosure in Plain English. Chicago: CCH Incorporated, 1999.
Greene, Anne E. Writing Science in Plain English. Chicago: University of Chicago Press, 2013.
Iverson, Cheryl, et al. AMA Manual of Style. 10th ed. Oxford: Oxford University Press, 2007.
Kimble, Joseph. Lifting the Fog of Legalese. Durham, NC: Carolina Academic Press, 2006.
Office of Investor Education and Assistance. A Plain English Handbook. Washington, DC: US
Securities and Exchange Commission, 1998.
Strunk, William, Jr. and E. B. White. The Elements of Style. 3rd ed. New York: Macmillan, 1979.
Stedman’s Medical Dictionary. 28th ed. Philadelphia: Lippincott Williams & Wilkins, 2006.
Tufte, Edward R. The Visual Display of Quantitative Information. 2nd ed. Cheshire, CT: Graphics
Press, 2001.
University of Chicago Press. Chicago Manual of Style. 15th ed. Chicago: University of Chicago
Press, 2003.
Williams, Joseph M. Style: Lessons in Clarity and Grace. 9th ed. New York: Pearson Longman, 2007.
Wydick, Richard C. Plain English for Lawyers. Durham, NC: Carolina Academic, 2005.
Articles
Gopen, George D., and Judith A. Swan. “The Science of Scientific Writing.” American Scientist 78
(November–December 1990), http://www.americanscientist.org/issues/pub/the-science-
of-scientific-writing/1.
Orwell, George. “Politics and the English Language,” Horizon (April 1946). Available online,
http://www.orwell.ru/library/essays/politics/english/e_polit/.
Internet References
Plain Language Action and Information Network. Federal Plain Language Guidelines. (March
2011, revised May 2011), www.plainlanguage.gov.
Plain Language at NIH, National Institutes of Health, https://www.nih.gov/institutes-nih/nih-
office-director/office-communications-public-liaison/clear-communication/plain-language.
205
206
â•‡ 207
INDEX
References to figures and tables are denoted by an italicized f and t
1066 principle, 73 Canada, English speaking, 118t, 120

Anglo-â•‰Saxon & Norman French words, 74t chart, two-â•‰dimensional data in, 96–â•‰98
applying to math, 74–â•‰75, 75t Chicago Manual of Style, 30–â•‰31
avoiding long Latin words to describe the real China, English speakers, 119t, 122
world, 79–â•‰82 Chrystal, David, 2
consistent use of terms, 78–â•‰79, clear, 5
minimizing false signal words, 80t, 80–â•‰81 closed compound, 30–â•‰32
preferring short word to describe real compound words
world, 75–â•‰76 closed compound, 30–â•‰32
showing real-â•‰world possession or hyphenated compound, 30–â•‰32
connection, 76–â•‰80 hyphenation of, 32–â•‰33
abstract open compound, 30–â•‰32
definition, 61 promoting reading ease, 29–â•‰33
examples of, 62t standard usage for, 30–â•‰31
identifying, 61–â•‰63 strategies for managing, 31–â•‰32
real world vs, 74t types of, 30
real world vs abstract possession or concise, 1, 4–â•‰5
connection, 77t concrete language
revising into concrete, 64–â•‰65 definition, 61
active voice, 49–â•‰59 favoring real-â•‰world adjectives, 66
converting nominalization into a verb in, identifying abstract and concrete
58–â•‰59 subjects, 61–â•‰63
definition, 49 nouns and verbs for meaning, 65–â•‰67
examples of, 50t one doctor treating one patient, See one doctor
identifying, 50–â•‰51 revising abstract into concrete, 64
minimizing to be and to have, 53–â•‰55 writing in the singular, 67–â•‰68
nominalization, 55–â•‰58 concrete terms, examples of, 62t
revising passive into, 51–â•‰52 connection, real-â•‰world vs abstract, 77t
vivid sentences, 54t content, 5. See also essential scientific content
The AMA Manual of Style (Iverson), 97 convention, challenging, 9
American Family Physician (journal), xv, 16 countries
and, starting sentence with, 36–â•‰37 English-â•‰speaking, 117–â•‰22
Anglo-â•‰Saxon words, 73, 74t top25 English-â•‰speaking, 118–â•‰19t
audience, widest reasonable, xi, xv, 2–â•‰5, The Craft of Research (Booth, Colomb and
9, 117 Williams), 97
Australia, English speaking, 118t, 120 Crick. See Watson and Crick
Bangladesh, English speaking, 118t, 120–â•‰21 Darwin, Charles, 107

British Medical Journal (journal), xv, 76 DeBakey, Selma, 1
but, starting sentence with, 36–â•‰37 dependent clause, 6t, 16, 23, 43
207
208
208 Index
Dick and Jane, 65–66 kaizen, improvement, 8–9

Dewey, John, 91 kidney transport, mathematical modeling
DNA. See Watson and Crick of, 125–26
Doyle, Sir Arthur Conan, 76 Kimble, Joseph, xiii, xvii, 2
Kübler-Ross, Elisabeth, 9
Eagleston, Robert, 2
Edison, Thomas, 78 The Lancet (journal), xv
editorial blindness, 3 language. See concrete language; vivid language
Egypt, English speaking, 118t, 120–21 Latin terms, sense of perception, 81t
Einstein, Albert, 85, 115 Lifting the Fog of Legalese: Essays on Plain Language
elegant variation, 78 (Kimble), xiii
The Elements of Style (Strunk and logical flow, xiv–xv
White), xii, 65 logical reasoning
English speakers, 117 explaining each step of, 107–8
China, 119t, 122 organizing the narrative. See narrative
comparing large English-speaking organization
countries, 122t owner's manual, 91–92
Europe, 121–22 stating the problem, 108–10
India, Pakistan, Nigeria, The Philippines, using words before symbols, 110–12
Bangladesh and Egypt, 118t, 120–21 long words, 27–39
top25 English-speaking countries, 117, definition, 27
118–19t, 120–22 essential scientific terms, 27
USA, UK, Canada and Australia, 118t, 120 See also words
essential scientific content, 5, 5f
essential scientific terms, 27–29 main point, 23–24
Europe, English speaking, 121–22 mathematical modeling of kidney
exercises, notes on, 10 transport, 125–26
Mayo Clinic Proceedings (journal), xv
false-signal words, 80–81 medical knowledge, 1
Flesch-Kincaid Grade Level tests, xiv, 13–15 Medical Library Association (Journal), 2
Flesch Reading Ease, xiii, xiv, 13–15 medical profession, teaching, 1
Franklin, Benjamin, 101 medical terms, citing common, 29
medicus incomprehensibilis, xi, 6–7, 85, 89
Garner, Bryan, xiii, xvii, 93 compound words, 29–33
Germany, 3 definition, xi
Goethe, Johann Wolfgang von, 61 diagnosing and treating, 15–18
grade level score, Flesch-Kincaid Grade Level, long words and, 37–38
13–15, 14t revising to reduce, 17t
graph, two-dimensional data in, 96–97 symptoms of, 6t, 6–7
Greene, Anne E., xiii, xvii, 27, 73 Michigan Bar Journal (journal), xiii, xvii
grief, stages of, 9
narrative organization
Hippocratic Oath, 1 one idea in each paragraph, 93
Hoffman, Hans, 41 two-dimensional data in table, chart or
Holmes, Sherlock, 76–77 graph, 96–97
The Hound of the Baskervilles (Doyle), 76–77 narrative pathway, 101–5
Hume, David, 47 anchoring discussion in real world, 102–3
hyphenated compounds, 30–32 choosing a good, 103–4
examples of common, 104t
India, English speaking, 118t, 120–21 starting with things known, 101–2
insider, xiv–xv, 2, 107 transition between concrete and abstract, 104–5
Nature (journal), 19
Jenner, Edward, 9 needless words
Journal of General Internal Medicine (journal), 2 of, 42–43
Journal of the American Medical Association omitting, 41–42
(journal), xv that, 43–44
Journal of the Medical Library Association The Netherlands, English speaking, 118t, 122
(journal), 2 New England Journal of Medicine (journal), xv
209
Index 209
Nigeria, English speaking, 118t, 120–21 reading time, 2

nominalization, 55–59 real-world facts, anchoring discussion
converting into verb in active voice, 58–59 in, 102–3
examples of, 56t Richard III (play), 53
identifying, 55–58
minimizing, 58t scientific terms, keeping essential, 27–29
when to use, 56–57 sentence, starting with and or but, 36–37
Noonan syndrome, 16, 17t sentence length
Norman French words, 73, 74t average and maximum, 19–22
nouns, concrete language, 65–67 common wisdom about, 20–22
noun strings keep subject and verb close together, 22–23
avoiding, 34–36 main point and commentary, 23–25
when to use, 35 reading ease, 6
using normal, 19–21
of, omitting, 42 writing tips, 7–8
On Death and Dying (Kübler-Ross), 9 Shakespeare, William, xiv, 53
one doctor treating one patient, 69–71 Sherlock Holmes, 76–77
open compound, 30–32 short word, 27
organization. See narrative organization signal, 80, 202
Orwell, George, 19 singular subject
owner's manual, scrambled, 91–92 revising to use, 68t
writing in, 67–69
pain medicine, describing dosing for, 112t one doctor treating one patient. See one
Pakistan, English speaking, 118t, 120–21 doctor
paragraph Smith, Dodie, 67
one idea, 93 stages of grief, 9
number of words, 93 Stedman's Medical Dictionary, xii, 31, 80
passive voice Strunk and White. See Elements of Style
definition, 47 Style: Toward Clarity and Grace (Williams), xiii
examples of, 50t subject, definition, 202
identifying, 50–51 symptoms. See medicus incomprehensibilis
when to use, 52–53
past participle, 50, 52, 55 table, two-dimensional data in, 96–97
perception, sense of, 81t that, omitting, 43–44
Philippines, English speaking, 118t, 120–21 Titanic, HMS, 110
plain English traditional medical writing, 15
challenging convention, 9
concepts of writing, xiv–xv UK (United Kingdom), English speaking,
description of, 4–6 118t, 120
essential scientific content, 5f United States of America (USA), English speaking,
how to use writing tips, 7–9 118t, 120
improving a journal's reading ease, 15f
reasons for writing in, 1–4 vasectomy reversal, 96–97, 97t
widest reasonable audience. See widest vasoepididymostomy, 96–97, 97t
reasonable audience vasovasostomy, 96–97, 97t
Plain English for Lawyers (Wydick), xii verbs
possession, real-world vs abstract, 77t concrete language, 65–67
pronunciation, hyphenation for, 31–32, 32t definition, 203
quality management, kaizen, 8–9 minimize to be and to have, 50, 53–54
vivid language, xiv, 47
reading ease, xiv, xv voice, definition, 49. See also active voice and
compound words promoting, 29–33 passive voice
Flesch-Kincaid Grade Level tests, 13–15
Flesch Reading Ease, 13–15 Watson and Crick, 9, 19
sentence length, 6, 19–21 widest reasonable audience, xi, xv, 2–5, 9, 117
taking charge of score, 17–18 William, Duke of Normandy, 73
word length, 27–28 Williams, Joseph, xiii, 49, 97
WSEG scores, 15–18 Wittgenstein, Ludwig, 13
210
210 Index
words omitting unnecessary endings, 33–34

and or but to start sentence, 36–37 percentage of long, 37
avoiding noun strings, 34–35 transition to symbols, 111–12
compound, 29–32 using, before symbols, 110–12
keeping essential scientific terms, 27–28 writing habits, overused, 6t
minimizing long, 27–28 Writing Science in Plain English
omitting needless, 41 (Greene), xiii, 27, 73
omitting needless of, 42 WSEG scores, 15–18
omitting needless that, 43–44 statistical analysis of, 85–89
211
212
213
214

Plain English For Doctors and Other Medical Scientists 2017 PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Plain English For Doctors and Other Medical Scientists 2017 PDF

Uploaded by

Copyright:

Available Formats

i

Plain English for Doctors

Plain English for Doctors

Published in the United States of America by Oxford University Press

© Oxford University Press 2017

All rights reserved. No part of this publication may be reproduced, stored in

You must not circulate this work in any other form

Library of Congress Cataloging-​in-​Publication Data

9 8 7 6 5 4 3 2 1

C . ME DICU S IN COMPR E HE N S IBIL I S 6

D. U SING T HE T IP S IN YOUR W R I T ING 7

C O N C E P T 1 ​ ■ TAKE CHARGE OF YOUR READING EASE SCORE

B. WSE G S C OR E S—​H O W W E T R ACK K EY R E A DING E A SE DATA 15

CHAPTER 1. Use normal sentence length 19

B. K E EP T HE S UBJE CT A ND V E R B CL O SE TO G E T HE R IN T HE FIR S T SEV E N

C . PU T T HE M A IN POIN T FIR S T A ND T HE N GI V E C OMME N TA RY,

CHAPTER 2. Prefer the short word 27

B. C I T E A C OMMON ME DIC A L T E R M ONC E TO AVOID C ONFU SION 29

D. OMI T A N Y UNNE C E S S A RY WOR D E NDING 33

E . AVOID T HE NOUN S T R ING 34

G . AVOID U SING A HIGH PE RC E N TAG E OF L ONG WOR DS 37

CHAPTER 3. Omit any needless word 41

C O N C E P T 2 ​ ■ USE VIVID LANGUAGE

CHAPTER 4. Prefer active voice 49

B. R EVI SE PA S SI V E IN TO ACT I V E VOIC E 51

C . W HE N SHOULD YOU U SE PA S SI V E VOIC E? 52

E . IDE N T IFY NOMINA L IZAT ION 55

F. C ONV E RT NOMINA L IZAT ION IN TO A V E R B IN ACT I V E VOIC E 58

CHAPTER 5. Prefer concrete language 61

B. R EVI SE ABS T R ACT IN TO C ONCR E T E 64

C . U SE NOUN S A ND V E R BS TO C A R RY T HE W E IGHT OF ME A NING 65

E . TA LK IN T E R MS OF ONE D O CTOR T R E AT ING ONE PAT IE N T 69

CHAPTER 6. Observe the 1066 principle 73

B. PR EF E R ’S TO SHO W R E A L-​W OR LD PO S SE S SION OR C ONNE CT ION 76

C . U SE T E R MS C ON SI S T E N TLY; AVOID E L E G A N T VA R I AT ION 78

D. AVOID U SING A L ONG , L AT IN WOR D TO DE S CR IBE T HE R E A L WOR LD 79

CHAPTER 7. Statistical analysis of wseg scores 85

CONCEPT 3​ ■ PRESENT LOGICAL REASONING CLEARLY

CHAPTER 8. Organize your narrative in a way that’s helpful for

B. PR E SE N T T WO-​D IME N SIONA L DATA IN A TABL E , C H A RT OR G R A PH 96

CHAPTER 9. Choose a clear narrative pathway 101

B. S TA RT BY A NC HOR ING YOUR DI S CU S SION IN T HE R E A L WOR LD 102

C . C HO O SE A GO OD NA R R AT I V E PAT HWAY 103

D. M AK E A SMO O T H T R A N SI T ION BE T W E E N C ONCR E T E A ND ABS T R ACT 104

CHAPTER 10. Forge a strong chain of logical reasoning 107

B. S TAT E T HE PROBL E M BEF OR E YOU S OLV E I T 108

C . S AY I T IN WOR DS BEF OR E YOU S AY I T IN SYMB OL S 110

Afterword—Can things ever change? 115

Appendix 1. English speakers around the world 117

Medicus incomprehensibilis is a condition that affects doctors and other medical

Who are we and how did we come to write this book?

How we structured the book

We thank and acknowledge several people for their contributions:

A. Why bother to write in plain English?

2 Plain English for Doctors and Other Medical Scientists

5. I T HE LP S YO U R WOR K R E AC H T HE W IDE S T

8 . I T S HO W S R E SPE CT F OR YOUR WOR K

9 . I T HE LP S OV E R C OME E DI TOR I A L BL I NDNE S S

As a medical researcher, your time is valuable. Learning to write and revise in

4 Plain English for Doctors and Other Medical Scientists

Exercise A. Widest reasonable audience

Compare your answers to the Exercise Key in Appendix 3.

B. What do we mean by plain English?

Library of Congress Cataloging-in-Publication Data

C O N C E P T 1 ■ TAKE CHARGE OF YOUR READING EASE SCORE

B. WSE G S C OR E S—H O W W E T R ACK K EY R E A DING E A SE DATA 15

C O N C E P T 2 ■ USE VIVID LANGUAGE

B. PR EF E R ’S TO SHO W R E A L-W OR LD PO S SE S SION OR C ONNE CT ION 76

CONCEPT 3 ■ PRESENT LOGICAL REASONING CLEARLY

CHAPTER 8. Organize your narrative in a way that’s helpful for

B. PR E SE N T T WO-D IME N SIONA L DATA IN A TABL E , C H A RT OR G R A PH 96

Figure I-1 Plain English is essential scientific content, presented clearly

Table I-1. Symptoms of medicus incomprehensibilis: over-used writing habits

Table I-2. Exercise C. Symptoms of medicus incomprehensibilis observed

A. Flesch Reading Ease and Flesch-Kincaid

Table C1-1. Flesch Reading Ease vs. Flesch-Kincaid Grade Level1

B. wseg scores—How we track key reading ease data

Example—Using reading ease scores to help diagnose and treat medicus

Table C1-2. Revising to reduce medicus incomprehensibilis

1. Although SDM is well accepted in overtly value-laden clinical deci-

Table 1-1. Put the main point first